CA2131003A1 - Novel cytokine that binds cd30 - Google Patents

Abstract

There is disclosed a polypeptide (CD30-L) and DNA sequences, vectors and transformed host cells useful in providing CD30-L
polypeptides. The CD30L-polypeptide binds to the receptor known as CD30, which is found on Hodgkin's Disease tumor cells.

Description

WO 93/24135 PCI/U~93/0492~

., 2~:10~)~

S ~TLE

NOV}~L CYTOKINE THAT-BINDS CD30 _ ~e.

Hodgkin's I~isease is a human Iymphoma, the etiology o~ which is still not well underslood. The neoplastic cells of Hodglcin's Disease are knvwn as Hodgl~n and Reed-Sternberg ~H-RS) cells. CD30 is a 120 kd sur~ace an~gen widely used as a clinical marker for Hodgkin~s Iymphoma and related herhatologic malignancies (Fr~ese et al., J. Immunol. 139:2081 (1987); Pfreundschuh et al., Onkologie 12:30 (1989~;
Carde et al., Eur. J. Cancer ~6:474 (1990)). Originally identified by the monoclonal antibody Ki- 1, which is reactive wi~h H-RS cells (Schwab e~ al., Nature (London) 299:65 (19~2)), CD30 was subsequen~ly shown to ~e expressed on a subset of non-Hodgkin's Iymphomas (N~), including Burkitt's Iymphoma, as well as several virally-transformed lines (human T Cell Lymphot~ophic vilus I or II transforrned T . ~ :
cells, and Epstein-Ba~r Virus transfonned B cells (Stein et ~.7 Bloocl 66:848 (1985);
Andreeson et al., Blood 63:1299 (1984)). Indeed, overall, 50% of Hodgkin's ~:
lymphomas are EBV~ ~Klein, Blood 80:299 (1992)). Tha~ (:D30 plays a role in ?5 nomnal:lvmphoid interactions is suggested by its histological de~ection on a small population of Iymphoid cells in reac~ive Iymph nodes, and by induced expression on purified T and B cells following lectin activation ~Stein et al., Int. J. Cancer 30:445 ~1982) and Stein et al., 1985, supr~).
Cloning and expression of a gene encoding CD30 has been reported and CD30 has been characteriz~d as a ~ansmembrane protein that possesses substan~ial homology ~ -to the nerve grc>wth factor receptor superfamily (Durkop et al., Cell 68:421, 1992). ~ -Durkop et al. suggest that CD30 is the receptor for one ~r more as yet unidentified :
grow~h factors, and recogniæ the importance of investigaang ~he existence and nature -of such g;rowth factors in order to achieve insight into the eiology of Hod~kin's 35 Disease.
Prior to the present invention7 however, no such growth factors or other molecules that bind to the CD30 receptor were known. A need thus remained for iden~ification and charac~erization of a ligand ~or CD30.

Wl~ ~3/24135 Pcr/U~93/o~926 ~3 2 The present invention provides a novel cytokine designated CD30-L, as well as isolated DNA encoding CD30-L protein, expression vectors comprising the isolated5 DNA, and a method for producing CD30-L by cultivating host cells containing the expression vectors under conditions appropriate for expression of the CD3~L protein.
CM0-L is a ligand that binds tO the Hodgkin's disease-associated antigen CD30 ~a cell surface r~ceptor). Antibodies directed against the CI:)30~1_ protein or an immunogenic fragment thereof are also provided.
lG
~ , .
Figures la and I b present a cDNA and encoded amino acid sequence for the receptor known as CD30. This sequence was reported by Durkop et al. (Cell 68:421, 1992). The signal peptide is underlined and the transmembrane region is designated by a double underline. ~:
Figure ~ presents cDNA and encoded amino acid sequences for a human lgG1 Fc fragmes~t. The Fc fragment was used to prepare a CD30/Fc fusion protein used in screening procedures tO isolate (~D30-L cDNA.
Figure 3 presents a DNA sequence, and the amino acid sequence encoded ~20~ thereby, for the coding region of a murine CD30-L cDNA clone, as described in Example 4. The transmembrane region is underlined. Nucleotides are numbered in the left margin: arnino acids in the nght margin.
Figure 4 presents a partial amino acid sequence for a human CD30-L cDNA
clone as described in Example 6. ~he human (h) sequence is aligned with an N-~5 ~ terminal portion of the murine (m) sequence (amino acids 1-130). The transmembrane region is underlioed~for ~he munne sequence and overlined for the human se~uence.
Figure ~ presents~a DNA sequence, and the amino acid sequence encoded thereby, ~or the coding region of a human CD3~L cDNA clone~ as des~ibed in Example 6. Thetransmembraneregionisunderlined. Nucleotidesarenumberedin the :

3 0 left margin; amino acids in the nght mar~in.
Figure 6 presents a DNA sequence, and the amino acid se~uence encoded thereby, for the coding region of a mu3ine CD30-L cDNA clone, as descnbed in Example 7. The transmembrane r~gion is underlined. The encoded protein compnses 19 addltional amino acids at ~e N-tem~inus when compa~ed with the sequence of figure 3. -Figure 7 presents a DNA sequence, and the amino acid sequence eneoded thereby, for the coding region of a human CD30-L cDNA clone7 as descnbed in Example 7. The transmembrane region is underlined. The encoded protein comprises ;

WO 93/24135 PCT/USg3/~49X6 19 additional amino acids at the N-terminus when compared with the sequence of figure 5.

S cDNA encoding a novel polypeptide that can act as a ligand for the Hodgkin's Disease-associated receptor known as CD3û has been isolated in accordance with the present invention. Also prwided are expression vectors complising the CD30 ligand (CD30-L) cDNA and methods for producing recombinant CD30-L polypeptides by cul~ivating host cells containing the expression vectors under conditions appropriate for expression of CD3~L9 and recovering the expressed CD30-L. Puriffe~ CD30-L
protein is also encompassed by the present invention.
The present invention also provides CD30-L or an~igenic fragments thereof that can act as immunogens to generate antibodies specific to the CD30-L immlmogens.
Monoclonal antibodies specific for CD30-L or anti~enic fragments thereof thus can be ~15 prepared.
The novel cytobne disclosed herein is a ligand for CD30, a receptor that is a member of the TNP/NGF recep~or super~amily. Therefore, CD30-L is likely to be responsible for transducing a biological signal via CD30, which is kDown to be expressed on the surface of Hodgkin's Disease tumor cells.
One use of the CD30 ligand of the present invention is as a research tool for studying the patho~enesis of Hodgkin's Disease. As described in example 8~ CD30-L
enhances the proliferation of the CD30~ neoplastic Hodgkin's Disease-derived Iymphoma cell line HDL~1-2. The HDLM-2 cells are pheno~rpically T-cell-like.
CD3~L did not pr~duce a detec~able effect on prolifera~ion or viability of the B-cell- ~;
~5 like, CD30+. Hodgkin's Disease-derived Iymphoma cell lines KM-H2 and L-428. The CD30-L of the present invention provides a means for investigating ~he roles that ;
CD30-L and the cognate~receptor may play in the etiolQgy of Hodgkin's Disease.
CD30-L exhibited a cytotoxic effect on the CD30+ non-Hodgkin's lymphoma eell line Karpas 299 (see example 0. Thus, CD3~L has poten~ial use as a therapeutic agent.
The CD3q ligand also induees proliferanon of T cells in the presence of an anti-CD3 c~s~imulus. The (~D30-L of the present invention ~hus is also useful as a research ~ool for elucidating the roles that CD30 and CD30-L may play in the irr~une syste~. The inducible exp~ession of CD30-L on normal T cells and ma~ophages, asld the presence of its receptor on activated T and B cells, is consisLent with both au~ocrine and paracrine effects. ~ ~
Upregulation of CD30 accompanying ~B~, HlLYI and HTLVII
transformanon also warrants fu~her inves~igation, and the CD30-L proYided herein is use~ s~ch studies. HTLVI is the proximal cause of adult T cell Leukemia/Lymphoma. EBV has long been associated with Burkitt's Iymphorna and nasopharyngeal carcinoma, and, overall, 50% of Hodgkin's Iymphomas are EBV+
(reviewed in Klein, 1992, supra). :
S The CD30-L polypeptides of the present invention also may be employed in in ~itro assays for detection of CD30 or CD~30-L or the interactions thereof. Additional cell types expressing CD30 may be iae~tified, for example.
The term "CD30-L" as used herein refers to a genus of polypeptides which are capable of binding ~D30. Human CD30-L is within the scope of the present invention, as are CD30-L proteins derived from other mammalian species. As used herein, theterm "CD30-L" includes membrane-bound proteins ~comprising a cytoplasmic domain,a transmembrane region, and an extracellular domain) as well as truncated proteins that retain the CD3~binding property. Such truncated proteins include, for example, soluble CD30-L compnsing only the extracellular (receptor binding) domain.
Isolation of a cDNA enc~ding murine CD3~L is descnbed in examples 1-4 below. A human CD30-Fc fusion protein was prepared as described in example 1 for: ~ ~ use In screening clones in a direct expression cloning procedure, to identify those expressing a protein that binds CD30.
Briefly, total RNA was isolated from a vi~ally transformed human T-cell line designated HUT-102, which has been described by Durkop et al., supra, and Poiesz et al. (PNAS USA 77:7415-19, 1980). First strand cDNA was prepared using the total RNA as template. DNA encoding the ex~acellular domain of human CD30 was amplified by polymerase chain reac~on (PCR) using primers based on the human CD30 s~uence published~ by Durkop et al., supra~, and the amplified DNA fragrnent was5~ isoIated. 4n expression vector comprising ~he CD30 extracellular domain DNA fused in-frame to the N-terminus of a human IgGl Fc region DNA sequence was constructed and transfected into rna~alian cells. The expressed protein was p~i~led by a procedure;that mvolved use of a protein G column (to which the Fc por~on of the fusion~protein binds).
Three activated munne helper T-cell lines were screened using a fluorescence -activated cell sorting technique, and all three were found to bind a fluorescent denvative of the (~D30-~c protein. A~cDNA library was prepared from one of the murine helper ~T-cell~lines. cDNA from this libra~ (in a mammalian expression vector ~at also replicates in E. coli) was transfected into COS-7 (man~nalian) cells9 ~or isolation of 35 ~ clones expressing a CD30-binding protein by using a direct expression cloning techn~que. The clones were screened~ for~ability to bind 1251-CD30/Fc, and a positive clone was isolatèd. The recombinant vector isolated from the positive clone (munne CD30-L cDNA in pl~smid pDC202) was transfonned into E. coli cells, deposited with : ' .

W~ 93/24135 PCI /U~93/04926 , .......................................................................... .
2~10~3 the American Type Culture Collection on May 28, 1992, and assigned accession no.ATCC 69()04. The deposit was made under the terms of the Budapest Treaty.
The murine CD30-L cDNA was radiolabeled and used as a probe tO isolate human CD30-L cDNA by cross-species hybridization. Briefly, a cDNA library 5 prepared from aceivated human peripheral blood Iymphocytes was screened with 32p labeled murine cDNA and a positive clone was isolated as desc~ibed in Example 6.Hurnan CD30-L DNA isolated from the positive clone was inserted into plasmid pCiEMBL and then transfonned into E. coli cells as described in Example 6. Samples of E. coli cells transforrned with the recombinant vector were deposited with the American Type Culture Collection on June 24, 1992, and assigned accession no.
ATCC 690?0. The deposit was made under the teIms of the Budapest Treaty.
Additional murine and human CD3~L DNA se~uences were isolated as described in example 7. The proteins encoded by the clones of example 7 compriseaddi~onal amino acids at the N-terminus, compared to the clones isolated in examples 4 ~15 and 6.
CD3()-L proteins of the present invention thus include, bu~ are not lirr~ited tO, :
murine CD3~-L proteins characterized by the N-term~nal arnino acid sequence Met-Gln-Val-Gln-Pro-Gly-Ser-Val-Ala-Ser-Pr~Trp (Figure 3) ~r Met-Glu-Pr~-Gly-Leu-Gln-Gln-Ala-Gly-Ser-Cys-Gly (Figure 6). Human CD30-L proteins characterized by the 20 N-terminal amino acid sequence Met-His-Val-Pro-Ala-Gly-Ser-Val-Ala-Ser-His-Leu (Figure 5) or Met-Asp-Pr~Gly:Leu-Gln-Gln-Ala-Leu-Asn-Gly-Met (Figure 7) also areprovided.
While a CD30/Fc fusion protein was employed in the screening procedure described in example 4 below, la~eled CD30 could be used to screen clones and candida~e cell lines for expression of CD30-L proteins. The CD3~/Fc fusion protein of~ers the advantage of being easily purified. In addi~on, disulfide bonds fonn be~ween the Fc regions of two separate ~usion protein chains, creating dimers. The dimeric CD30/Fc receptor was chosen for the potential advantage of higher affinity binding of the CD30 ligand, in view of the possibili~y that the ligand being sought 30 would be multimenc.
Fur~er, other suitable fusion pro~eins comprising CD30 may be substituted for CD3û/Fc in the screening procedures. Other fusion proteins can be made by ~using a DNA sequence for the Iigand binding domain of CD3û to a DNA sequence encoding another polypepude that is capable of affinity purification~ for example, avidin or 35 streptavidin. The resultant gene construct can be in~oduced into mammalian cells to express a fusion protçin. Receptor/avidin fusion proteins can be pun~led by bio~n ~ffinity chromatograpny. The fusion protein can later be recovered from the column by eluting with a high salt solution or another appropriate buffer. Other antibody Fc W~ 93/24135 ~ o~3 PCI/US93/04926 regions may be substituted for the human IgGI Fc re~ion described in example 1.
Other suitable Fc regions are defined as any region that can bind with high affinity to protein A or protein G, and include the Fc region of murine IgGl or fragments of the human lgG 1 Fc region, e.g., fragments comprising at least the hinge region so that 5 interchain disulfide bonds will fo~m.
cDNA encoding a CD30-L polypepude may be isolated f~m other mammalian species by procedures analogous to thQse employed in isolating the murine CD30~Lclone. For example, a cDNA libra~y derived from a different mammalian species may be substituted for the murine cDNA library that was screened for binding of radioiodinated human CD30/Fc fusion protein in the direct expression cloning -~
procedure described in example 4. Cel~ types from which cDNA libraries may be prepared may be chosen by the FACS selection procedure described in example 2, or any ~ther suitable technique. As one alternative. mRNAs iso~ated from valious cell Iines can be screened by i~lorthern hybridization to determine a suitable source of mammalian CD30-L mRNA for use in cloning a CD31)-L gene.
Alternatively, one can utilize the murine or human CD30-L cDNAs described herein to screen cDNA derived from vther marnrnalian sources for CD30-L cDNA
using cross-species hybridization techniques. Briefly, an oligonucleotide based on the ~nucleo~ide sequence of Ihe coding region (preferably the extracellular region) of the 20 murine or hurnan clone, or, prefer~bly, the full length CD30-L cDNA, is prepared by standard techniques for use as a probe. The murine or human probe is used to screen a marnmalian cDNA library or genomic library, generally under moderately stringent conditions.
CD30-L proteins of the present invention include, but are not limited to. murine~5 CD30-L comprising amino acids 1-220 of figure 3 or 1-239 of figure 6; human CD3 L comprising amino acids 1-215 of figure 5 or 1-234 of figure 7; and proteins that compnse N-terminal, C-terminal, or internal trunca~ons of ~he foregoing sequences, but re~in ~he desired biological ac~vity. Examples include murine CD3~L proteinscomprising amino acids x to 239 of figure 6, wherein x is 1 19 (i.e., the N-terminal amino acid is selected from amino acids 1-19 oÇ figure 6, and the C-terminal amino acid is amino acid 23g of figure 6.) As described in example 7, amino acids 1-l9 of the figure 6 sequence are not essendal for binding of rnurine CD3()-L to the CD30 recept~r.
Also prvvided by the present inven~ion are human CI)30-L proteins comprising amino acids y to 234 of figure 7 wherein y is 1-19 (i.e., the N-terminal amino acid is any one of a~no aeids 1-19 of figure 7, and amim~ acid 234 is the C-terminal a~T~no acid.
Such p~teins, truncatcd at the N-~erminus, are capable of binding CD30, as diseussed in example 7.

WO 93/2413~ PCr/U~3/0~926 7 21~10~3 One embodiment of the present invention provides soluble CD3~L
polypeptides. Soluble CD30-L polypeptides comprise all or part of the extracellular domain of a native CD30-L but lack the transmembrane region that would cause retention of the polypeptide on a cell membrane. Since the CD30-L protein lacks a S signal pep~ide, a heterologous signal peptide is fused to the N-terminus of a soluble CD30-L protein to promote secretion thereof, as described in more detail below. The signal peptide is cleaved frcJm the CD3~L protein upon secretion from the host cell.
The soluble CD3~L polypeptides that may be employed retain the ability to bind the CD30 receptor. Soluble CD30-L may also include part of the transmembrane region or 10 part of the cytoplasmic domain or other sequencest provided that the soluble CD3~L
protein is capable of being secreted.
Soluble CD30-L may be identified (and distinguished from its non-soluble rnembrane-bound counterparts) by separating intact cells which express the desired proteiri from Ihe culture medium, e.g.~ by centrifugation, and assaying the medium (supematant) for the presence of the desired protein. The culture medium may be assayed using procedures which are similar or identical to those described in the examples below. The presence of CD30-L in the medium indicates that the protein was secreted from the cells and thus is a soluble ~onn of the desired protein.
The use of soluble fonns of CD3~L is advantageous for certain applications.
20 Puri~lcation of the proteins from recombinant host cells is ~acilitated, since the soluble - p~oteins are secreted from the cells.
Examples of soluble CD3~L polypeptides include those comprising the entire extracellular domain of a naaive CD30-L protein. One such soluble CD3~L comprises amino acids 49 (Gln) through 220 (Asp) of the murine CD30-L sequence of Figure 3.
25 Other soluble CD3~L polypeptides comp~ise amino acids z tO 215 (Asp) of the human CD30-L sequence of Figure 5, wherein z is 44, 45, 46, or 47. Xn other words, the N-tenninal amino acid of the soluble human CD30-L is selected from the amino acids in positions 44-47 of Figure 5. ~
Truncated M0-L, including soluble polypeptides, may be prepared by ~ny of 30 a number of conventional teehniques. ln the case ~f recombinant proteins, a DNA
fragment encoding a desired fragmçnt may be subcloned into an expression vector.Alternatively, a desired DNA sequence may be chemically synthesized using known techniques. DNA fragments also may be produced by res~iction endonuclease digestion of a full length cloned DNA sequence, and isolated by elec~ophoresis on 35 agarose gels. Linkers containing res~iction endonuclease cleavage si~e(s) may be emp}oyed to insert the desired DNA fragment into an expression vector, or the fragment may be digested at cleavage sites naturally present therein. The well known polymerase :.

?, - PCr/U5~3/Oq926 chain reaction proeedure also may be employed to isolate a DNA sequence encoding a desired protein fragment.
In another approach, enzymatic treatment (e.g., using Bal 31 exonuclease) may be employed to delete terrninal nucleotides from a DNA fragrnent to obtain a fragrnent S having a palticular desired terminsls. Among the commercially available linkers are those that can be ligated to the blunt ends produced by Bal 31 digestion, and which contain restriction endonuclease cleavage site~s). Alternatively, oligonucleo~des that reconstruct the N- or C-terminus of a I)NA fragrnent to a desired point may be -synthesized. The oligonucleotide may contain a restriction endonuclease cleavage site upstream of the desired coding sequence and position an initiation codon (ATG) at the N-terminus of the coding sequence.
The present invention provides purified CD3~L polypeptides, both recombinant and non-recomb1nant. Variants and derivatives of native CD30-L proteins that retain the desired biological activity are also within the scope of the present 'I S invention. CD30-L variants may ~e obtained by mutations of nucleotide sequences coding for native CD30-L polypeptides. A CD30-L variant, as re~elTed to herein, is a polypeptide substantially homologous to a native CD30-L, but which has an amino acid sequence different from that of native CD30-L (human, munne or other mammalian species) because of one or a plurality of deletions, insertions or substitutions.
The variant arnino acid sequence preferably is at least 80% iden~ical to a native CD30-L amino acid seauence, most preferably at least 90% identical. The degree of homology (percent identity) may be determined, for example, by comparing sequence information using the GAP computer program, version 6.0 described by Devereux etal. (Nucl. Acids Res. 1~:387, 1984) and available from the University of Wisconsin ~; ~5 Genetics Çomputer Group (UWGCG). The GAP program utilizes the alignment method of Needleman and Wunsch (J. Mol. Biol. 48:443, 1970), as revised by Smithand Waterman (A~v. App~. Math 2:482, 1981). The prefelTed default parameters forthe GAP pr~gram include: ~1) a unary compar~son matrix (containing a value of 1 for identities and 0 for non-identi~es) f~r nucleotides, and the weighted comparison matrix of Gnbskov and Burgess, Nucl. Acids Res. 14:6745, 1986, as described by Schwartzand Dayhoff, eds., Atlas of Profein Sequence and Structure, Na~onal Biomedical Research Foundation, pp. 353-358, 1979; (2) a penalty of 3.() for each gap and an additional 0.10 penalty for each symbol in each gap; and (3) no penalty for end gaps.
Alterations of the native amino acid sequence may ~ accomplished by any of a number of known techniques. Mutations can be in~oduced at par~cular loci by synthesizing oligonucleo~ides containing a mutant sequence, flanked by restriction sites enabling ligalion to fragments of the native sequence. F~llowing ligation, the resulting ":

WO g3/24135 2 i ~ 3 Pcr/US93/0492~

reconstructed sequence encodes an analog having the desired amino acid inser~on,substitution, or dele~ion.
Akernatively, oligonucleotide-dire~ted site-specific mutagenesis procedures can 'oe employed to provide an altered gene having particular codons altered according to S the substitution, deletion, or insertion required. Exemplary methods of making such alterations are disclosed by Walder et al. (Gene 42:133, 1986~; Bauer et al. (Cene 37:73, 1985); Craik (BioTechniques, January 1985, 12-19); Smith et al. (Genetic E;ngineering: Prfnciples and Methods, Plenum Press, 1981); and U.S. Patent Nos.

4,51g,584 and 4,737,462, which are incorporated by reference hcrein.
1 O Variants rnay comprise conservatively substituted sequences, meaning that a given amino acid residue is replaced by a residue having similar physiochemical characteristics. Examples of conservative substitutions include substitution of one aliphatic residue for another, such as lle, Yal, Leu, or Ala for one another, orsubstitutions of one polar residue for another, such as between Lys and Arg; Glu and Asp; or Gln and Asn~ Other such conselvative substitutions, for example, substitu~vns of ennre regions having similar hydrophobicity characteristics, are well kr own.CD30-L also may be modified to cr~ate CD30-L derivatives by forming cvvalent or aggregative conjugates with o~her chemical moieties, such as glycosyl groups, lipids, phosphate, acetyl groups and the like. Covalent denvatives of CD3~L
2~ may be prepared by linlcing the chemical moie~ies to functional g~oups on CD30-L
arn~no acid side chains or at the N-terminus or C-terrninus of a CD3~1, polypep~de or the ex~acellular domain thereof. Other desivatives of CD30-1, within the scope of this invenuon include covalent or aggregative conjugates of CD3~L or i~s fragments with other proteins or polypepddes. such as by synthesis in recombinant culture as N-~5 ~e~minal or C-tenninal~fusions. For example, the conjugate may comprise a signal or leader polypeptide sequence (e.g. the -factor leader of Saccharomyces) at the N-terminus of a soluble CD3~L polypeptide. The signal or leader peptide co-~ansla~ionally or post-~ansla~ionally di~ects transfer of the conjugate from itS site of synthesis to a site inside or outside of the eell membrane or cell wall.
CD3~L polypep~ide fusions can compnse peptides added to facilitate purifica~on and identification of CD30-1,. Such peptides include, for example, poly-His ~r the antigenic identification peptides described in U.S. Patent No. 5,011,912 and in Hopp et al., BiolTecknology 6:1204, 1988. One such peptide is the F~AG(~
peptide, Asp-Tyr-Lys-Asp-Asp-Asp-As~Lys (DYKDDDDK), which is highly antigenic and provides an epitope reversibly bound by a specific monoclonal antibody enabling rapid assay and facile puri~lcation of expressed recombinant protein. This sequence is also specifically cleaved by bovine mucosal enterokinase at the residue itrnmediately following the Asp-Lys pairing. Fusion proteins capped witù this peptide WO g3/24135 ~ PCr/US93/0~9~6 may also be resistant to intracellular degradation in E. coli. A murine hybridoma designated 4El 1 produces a monoclonal antibody that binds the peptide DYICDDD~Kin the presence of certain divalent metal cations (as described in U.S. Patent 5,011,912) and has been deposited with the American Type Culture Collection under accession no S HB 9259.
The present invention fur~her includes CD30-L polypeptides with or without associated native-pattem glycosylation. CD3~L expressed in yeast or mammalian expression sys~ems ~e.g., COS-7 cells) may be similar to or significantly different from a native CD3~L polypep~ide in molecular weight and glycosylation pat~tern~
10 depending upon the choice of expression system. Expression of CD30-L polypeptides in bac~erial expression systems, such as E. coli, provides non-glycosylated molecules.
DNA constructs that encode various additions or substitutions of amino acid residues or sequences, or deletions of terminal or internal residues or sequences not needed for biological activity or binding c~n be prepared. For example, ~J-glycosylation sites in the CD30-L extracellular domain can be modified to preclude glycosylation while allowing expression of a homogeneous, reduced carbohydrate analog using yeas~ or mammalian expression systems. N-glycosylation sites in euka~yotic polypep~ides are characterized ~y an amino acid triplet Asn-X-Y, wherein X
is any amino acid except Pro and Y is Ser or Thr. Appropriate modifications to the 20 nucleotide sequence encoding this triplet will ~esult in subs~itutions, addi~ons or deletions that prevent attachment of carbohydrate residues at the Asn side chain.
Alteration of a single nucleo~ide, chosen so that Asn is replaced by a different atnino acid, for example, is sufficient to inactivate an N-glycosylation site. Known procedures for inactivating N-glycosylation sites in proteins include those described in ~5 U.S. Patent ~,071,972 and EP 276,846.
ln another exampie, sequences encoding Cys residues that are not essential for biological activity can be altered to cause the Cys residues to be deleted or replaced with other amino acids, preventing fomlaaion of incorrect intramolecular disulfide bridges upon renaturation. Other variants are prepared by modification of adjacent dibasic 30 an~ino acid residues to enhance expression in yeast systems in which KEX2 protease activity is present. EP 212,914 discloses the use of site-specific mutagenesis to inactivate KEX2 protease processing sites în a pro~ein. KEX2 protease processingsites are inactivated by deleting~ adding or substituting residues tO alter Arg-Arg, Arg-Lys, and Lys-Arg pairs to eliminate the occuIIence of these adjacen~ basic residues.
3~ Lys-Lys pairings are considerably less suscepible tQ KEX2 cleavage, and conversion of Arg-Lys or Lys-Arg to Lys-Lys represents a conservanve and prefer~d approach to inactivaing K~X2 sites. The resulting muteins are less susseptible to cleavage by the WO93/~413~ ~ ~ 3 ~ 003 PCI/US93/04926 Il KEX2 protease at locations other than the yeast c~-factor leader sequence, wherecleavage upon secretion is intended.
Naturally occurring CD30-L variaslts are also encompassed by the present inven~on. Examples of such variants are proteins that result from alternative nRNA

5 splicing events (since CD30-L presumably is encoded by a multi-exon gene) or frorn proteolytic cleavage of the CD30-L protein, wherein the CD30-binding property isre~ned. Alternative splicing o~ mRNA may yield a truncated but biologically active CD3~L protein, such as a naturally occurring soluble fo~m of the protein, for example.
Variations attributable to proteolysis include, for example, differences in the N- or C-10 terrnini upon expression in different types of host cclls, due to proteolytic removal ofone or more terminal amino acids from the CD30-L protein (generally from 1-5 termin31 amino acids).
Nucleic acid sequences within the scope of the present invention include isolated DNA and RNA sequences ~hat hybridize to the CD3~L nucleotide sequences ~fs disclosed herein under conditions of moderate or severe stringency, and which encode biologically active CD30-L. Moderate stringency hyb;idization conditions refer to conditions described in, ~or example, Sambrook et al. Molecular Cloning: A
Laboratory Manual, 2 ed. Vol. 1, pp. 1.101-104, Cold Spring Harbor Labora~ory Press, tl989). Conditîons of moderate stringency, as defined by Sambrook et al.,include use of a prewashing solution of 5 X SSC, O.S% SDS, 1.0 mM EDTA (pH 8.0) and hybridization conditions of about 55C, S X SSC, overnight. Conditions of severe stringency include higher temperatures of hybridiza~ion and washing. The skilledartisan will recognize that the temperanlre and wash solu~on salt concentra~ion may be adjusted as necessary accordin~ to factors such as the length of the probe.
~5 The present invention thus provides isolated DNA sequences encoding biologically ac~ive CD30-L, sel~cted from: ~a) DNA denved from the coding region of a na~ive marnmalian CD3~L gene (e.g., DNA comprising the nucleotide sequence presented in figures 3, 5, 6, or 7; ~b) DNA capable of hybndizanon to a DNA of (a) under moderately stringent condi~ons and which encodes biologically acnve CD30-L;
and (c) DNA which is degenerate as a result of the gene~ic code to a DNA defined in (a) or (b) and which encodes biologically active CD30-L. CD30-L proteins encoded by the DNA sequences of (a), (b) and (c) are encompassed by the present invenno~.
Examples o~ CD30-L proteins encoded by DNA that varies ~rom the native DNA sequences of Figures 3, 5, 6, and 7, wherein the vanant DNA will hybridize to a na~ive DNA sequence under moderately s~ingent conditions, include, but are not ed to, CD30-L fragments (soluble or mernbrane-bound~ and CD30-L proteins comprising inac~ivated N-glycosylation site(s~, inactivated KEX2 protease processing site(s), or conserva~ive amino acid subs~itu~ion(s), as descnbed above. CD30-L

prot encoded by DNA derived from o~her malT~nalian species, wherein the DNA
will hybridize to the human or murine DNA of Figures 3, 5, 6, or 7, are also encompassed.
Variants possessing the requisite ability to bind CD30 may be identified by any 5 suitable assay. Biological activity of CD30-L may be determined, for example, by competinon for binding tO the ligand binding domain of CD30 (i.e. compentive binding zssays ~.
One type of a competitive binding assay for CD30-L polypeptide uses a radiolabeled, soluble human or murine CD30-L and intact cells expressing cell surface CD30 (e.g., cell lines such as Hl~T102, described by Durkop et al., supra). Instead of intact cells, one could substitute soluble CD30 bound to a solid phase ~such as a CD301Fc fusion protein bound to a Protein A or Protein G column through interaction with the Fc region of the fusion protein). Another type of competitive binding assay utilizes radiolabeled soluble CD30 such as a CD30/Fc fusion protein, and intact cells expressing CD30-L. Altema~ively, soluble CD30-L could be bound to a solid phase.Competitive binding assays can be performed using standard methodology.
For example, radiolabeled murine CD30^L can be used to compete with a putative CD30-L homolog to assay for binding activity against surface-bound CD30.
Qualitative results can be obtained by compe~itive autoradiographlc plate binding .
20 assays, or Scatchard plots may be utilized to generate quanti~ative results.
Competi~ive binding assays with intact cells expressing CD30 can be perforrned by two methods. In a first method, cells expressing cell su~face CD30 are grown either in suspension or by adherence to tissue culture plates. Adherent cells can be removed by treatment with S rnM EDTA treatment for ten minutes at 37u C. In a second method, ~5 transfected C(3S cells expressing membrane-bound CD3V can be used. COS cells or another mamrnalian cell can be ~ansfected with human CD30 cDNA in an appropliatevector tO express full length CD30 with an extracellular re~ion.
Alternatively, soluble CD30 can be bound to a solid phase such as a column chromatography ma~ix or a similar substrate suitable for analysis ~or the presence of a 30 detectable moiety such as 125I. Binding to a solid phase can be accornplished, for example, by obtaining a CD30/Fc fusion protein and binding it to a protein A or protein G-containing ma~ix.
Another means ~o measure the biological activity of CD30-L (including variants) is to u~ilize conjugated, soluble CD30 (f~ example, 125I~ 01Fc) in competition35 assays sin~ilar to those described ab~Ye. In this case, however, intact cells expressing CD3~L, or soluble CD30-L bound to a solid substrate, are used to measure competi~ion for binding of labeled, soluble CD3Q to CD30-L by a sample containing a ~ -putauve CD30-L variant. ~ -WO 93~ 35 PCI/US93/04~26 13~3 1 ~ ~3 The CD30-L of the present invention can be used in a binding assay to detect cells expressing CD30. For example, CD3~L or an ex~acellular domain or a fragrnent thereof can be conjugated to a détectable moiety such as 125I. Radiolabeling with l251 can be perfoImed by any of several standard methodologies that yield a funcdonal l25I-S CD30-L molecule labeled to high specific activity. Alternatively, another de~ectable moiety such as an enzyme that can catalyze a colorometric or fluorome~ic reaction, biotin or avidin may be used. Cells to be tested for CI:)30 expression can be contacted wi~h conjugaled CD30-L. After incubation, unbound conjugated CD30-L is removed and binding is measured using the detectable moiety.
CD30-L polypep~des may exist as oligomers, such as dimers or tnmers.
Oligomers may be linked by disulfide bonds formed between cysteine residues on different CD30-L polypeptides. In one embodiment of the invention, a CD30-L dimer is created by fusing CD30-L to the Fc region of an antibody (IgG1) in a manner that does not jnterfere with bmding of CD30-L to the CD30 ligand binding domain. The Fc polypeptide preferably is fused to the N-tenninus of a soluble CD30-L (comprising only the extracellular domain). A procedure for isolating DNA encoding an IgG 1 Fc region for use in preparing fusion proteins is presented in example 1 below. A gene fusion encoding the CD30-L/Fc fusic;n protein is inserted into an appropriate expression veceor. The CD30-L/Fc fusion proteins are allowed to assemble much like 2 0 antibody molecules, whereupon interchain disulfide bonds fo)m between Fc , polypeptides, yielding divalent CD30-L. If fusion proteins are made with both heavy and light chains of an antibody, it is possible to form a C1~30-L oligomer with as many as four D30-L extracellular regions.
Alternatively, one~can link multiple copies of CD30-L via peptide linkers. A
fusion protein compr~sing~two or more copies of (,'D3~L ~preferably soluble CD30-L
polypeptides), sep~rated by peptide linkers, may be produced by recombinant DNA
t echnology. Among the peptide linkers that may be employed are amino acid chains ~: ~ that are from S to 100 amino acids in lengdl, preferably comprising alT~no acids selected from the group consisting o~ glycine? asparagine, serine, threonine, and alanine. In one embodiment of the present inven~ion, a fusion protein comprises two or th~e soluble CD30-L polypeptides linked via a peptide linker selected from Cily4SerGlysSer and (Gly4Ser)n, wherém n is 4-12. The production of recombinant fusion proteins comprislng pep~de linkers is illus~ated in United States Patent 5,073,627, for example, ~
The present invendon provides oligomers ~f CD3~L extracellular domains or ; ~-fragmen~s thereof, linked by disulfide bonds, or exp~essed as ~usion proteins with or without spacer amin~ acld linking groups. For ex~nple, a dimer CD30-L molecule ean - be linked by an IgG Fc t~Egion linking gtoup. Analysis of expressed recombinant WO 93/24135 PC~/U~93/~4926 ~3~ C, CD30-L of the present invention by SDS-PAGE revealed both monomeric and oligomeric forrns of the protein. The CD30-L proteins of the present invention are ieYe~i tO form oligomers (disulfide-bonded dimers, trimers and highe~ oligomers)in~acellular~y. The oligomers then become a~ached to the cell surface via the S transmembrane region of the protein.
The present invention ~r~iidés recombinant expression vectors for expression of CD30-L" and host cells transformed with the expression vectors. Any suitable expression system may be employed. The vectors include a CD30-L DNA sequence (e.g., a synthetic or cDNA-derived DN~ sequence enc~ding a CD3~L polypeptide) operably linked to suitab}e transcriptional or translational regulatory nucleo~de sequences, such as those derived from a mammalian, n~icrnbial, viral, or insect gene.
Examples of regulatory sequences include transcriptional promoters, operators, or enhancers, an mRNA ribosomal binding site, and appropriate sequences which control tlanscription and translation initiation and terrnination. Nucleotide sequences are 'i S operably linked when Ihe regulatory sequence functionally relates tO the CD30-L DNA
sequence. Thus, a promoter nucleotide sequence is operab}y linked to a CD3~L DNAsequellce if the promoter nucleotide sequence controls the transcription of the CD3~L
DNA sequence. The ability tO replicate in the desired host cells, usually conferred by an origin of replication, and a selection gene by which transfonnants are identified, may additionally be incorporated into the expression vector.
ln addition, sequences encoding appropriate signal pep~ides tha~ are not native to the CD30-L gene can be incorporated into expression vectors. For example, a DNA
sequence for a signal peptide (secretory leader) may be fused in frame to the CD3~L
sequence so that the CD30-L is ini~ially translated as a fusion protein comprising the ~5 signal peptide. A signal peptid~ fused to the N-ter~inus of a soluble CD3û-L protein ~promo~es extracellular secre~ion of the CD3~L. The signal peptide is cleaved from the 1:)30-L polypeptide upon secretion of CD3~L from the cell. Signal pepndes are chosen according tv the intended host cells, and representative examples are described below.
Suitable host cells for expression of CD30-L polypeptides include prokaryotes, yeast or higher eukaryotic cells. Appropriate cloning and expression vectors for use with bacteaial~ fQIngal, y~ast7 and mammalian cellular hosts are descnbed, ~or example, in Pouwels et al. Cloning Vectors: A Laboratory Manual, Elsevier, New York, (1~85). ~ell-free ~anslation systems could also be employed to produce CD3~L
polypeptides using RNAs derived ~rom DNA constructs disclosed he~eîn.
Proka3yotes indude gram negative or gram positive organisms, for example7 E.
coli orBacilli. Suitable proka~yo~ic host cells for ~ansfonnation include, for example, E. coli, Bacillus subtilis. Salmonella ~yphimurium, and VarlOllS other species within the WO g3/24135 PCr/l~S93~04926 1S 2~3~ Q~3 genera Pseudosnonas, Streptomyces, and Staphylococcus. In a prokaryotic host cell, such as E. coli, a CD30-L polypeptide may include an N-te~ninal me~hionine residue to facilitate expression of the recombinant polypeptide in the prokaryohc host cell. The N-terminal Met may be cleaved from the expressed recombinant CD30-L polypeptide.
S Expression vectors for use in prokaryotic host cells generally compnse one or more phenotypic selectable marker genes. A phenotypic selectable marker gene is, for example, a gene encoding a protein that confers antibiodc resistance or that supplies an autotrophic requirement. Examples of useful expression vectors for prokaryotic host cells in~lude those derived from co~nercially aYailable plasmids such as the cloning vector pBR322 (ATCC 37017). pBR322 contains genes for ampicillin and te~acyclineresistance and thus provides simple means for iden~ifying ~ansformed cells. An appropriate promoter and a CD30-L DNA seqllence are inserted into the pBR322 vector. Other commercial}y available vectors include, for example, pKK223-3 (Pharinacia Fine Chemicals, Uppsala. Sweden) and pGElVI 1 (Promega Biotec.
Madison, Wl, USA).
Promoter sequences commonly used for recombinant prokaryotic host cell expression vectors include ~-lactamase (penicillinase), lactose promoter systèm (~hang et al., Nature 275:615~ 1978; and Goeddel et al., Nature 281:544, 1979)9 tryptophan (~p) promoter system (G~ddel et al., Nucl. Acids Res. 8:4057, 1980; and EP-A-36776~ and tac promoter (Maniatis, Molecular Cloning: A Laboratory Manual, Cold Spring llarbor Lab~ratory, p. 412, 1982). A particularly useful prokalyonc host cell expression system employs a phage ~. PL p~omoter and a cI857ts thermolabile repressor sequence. Plasmid vectors available fronn the American Type Culture Collection which incorporate delivatives of the ~ PL promoter include plasn~id pHlJB2 ~resident in E.
coli strain JMB9 (ATCC 37092)) and pPLc28 (resident in E. coli RRl (ATCC
53082)).
CD30-L alternatively may be expressed in yeast host cells, preferably from the Saccharomyces genus (e.g., S. cerevisiae). Other genera of yeast, such as Pichia or ::
:: Klu~veromyces, may also be employed. Yeast vectors will o~ten contain an origin of replication sequence fr~m a 2,u yeast plasmid, an autonomously replicating sequence (ARS), a promoter reglon, sequences fvr polyadenylation, sequences for ~anscrip~on tennination, and a selectable marker gene. Suitable promoter sequences fo~ yeastvectors include, among others, p~omoters for metallothionein, 3-phospho~lyceratekinase (Hitzeman et al., J. Biol. Chem. 255:2073, 1980) or o~er glycolytic enzymes .
(Hess et al., J. Adv. Enzyme Reg. 7:149, 1968; and Holland et al., Biochem. 17:4900, 1978), such as enolase, glyceraldehyde-3-phosphate dehydrogenase, hexokinase, pyruvate decarboxylase, phosphofructokinase, glucose-6-phosphate isomerase, 3-phosphoglycerate mutase, pyruvate kinase, ~iosephosphate isomerase, WO 93/24135 Pcr/US93/~4926 a~3 16 phosphoglucose isomerase, and glucokinase. Other suitable vectors and promoters for use in yeast expression are further described in Hitzeman, EPA-73,657. Ano~her alternative is the ~lucose-repressible ADH2 promoter described by Russell et al. (J.
Biol. Chem. 258:2674, 1982) and Beier et al. (Nature 300:724, 1982). Shu~tle S vectors replicable in both yeast and E. coli may be constructed by inserting DNA
sequences from pBR322 for selécXon and replication in E. coli (Ampr gene and origin of replication~ into the above-described yeast vectors.
The yeast o~-factor leader sequence may be employed to direct secretion of the CD30-L polypeptide. The cl-factor leader sequence is often inse~ted between the 10 prornoter sequence and the structural gene sequence. See, e.g.7 Kurjan et al., Cell 30:933, 1982; Bitter et al., Proc. Natl. Acad. Sci. USA 81:5330, 1984; U.S. Patent 4,546,082; and EP 324,274. Other leader sequences suitable for facilitating secretion of recombinant polypepti~es from yeast hosts are known to those of skill in the art. A
leader sequence may be modified near its 3' end to contain one or more restriction sites.
.. ~ . . .
This will facilitate fusion of the leader sequence tO the structural gene.
Yeast tlansformation protocols are known to those of skill in the art. One such protocol is described by Hinnen et al., Proc. Natl. Acad. Sci. USA 75:1929, 1978.
The Hinnen et al. protocol selects for Trp~ transformants in a selective medium,wherein the selective medium consists of 0.67% yeast ni¢ogen base, 0.5% casaminoO acids, 2% glucose, 10 I,lg/ml adenine and 20 llg/ml uracil.
Yeast host cells transformed by vectors containing ADH2 promoter seguence may be grown for inducing expression in a "rich" medium. An example of a rich medium is one consis~ing of 1% yeast ex~act, 2% peptone, and 1% glucose supplemented with 80 ~lg/ml adenine and 80 llg/ml UMCil. Derepression of the ADH~
~5 promoter occurs when glucose is exhausted from the medium.
Mammalian or insect host cell culture systems could also be employed tO
express~ recomb}nant CD30-L polypepddes. Bacu!ovirus systems for produc~ion of heterologous proteins in insecl cells a3e reviewed by Luckow and Summers, BiolTechno~ogy 6:47 ~198~8). Es~ablished ~ell lines ~ mammalian origin also may be empl~yed. Examples of suitable ma~alian host cell lines include the COS-7 line of monkey l~idney cells (ATCC CRI, 1651) (Gluzman e~ al., Cell 23:175, 1981), L cells, (~127 cells, 3T3 cells (ATCC CCL 163), Chinese hamster ova~y (CHO~ cells, HeLa cells, and BHK (ATCC (~RL 10~ cell lines. and the CVl/EBNA cell line derived from the Afncan ~reen monkey~ kidney cell line CV l (ATCC CCL 70) as described by McMahan et al. ~EMBO J. 10: 2821, 1991).
Transcnptional and ~ansla~ional con~ol sequences for mammalian host cell expression vectors may b~ excised from viral genomes. ~ommonly llsed promoter sequences and enhancer sequences are derived from Polyoma virus, Adenovirus ~, ~ ..

WO 93/2413~ PCI /US93/~4926 17 ~ 3 Simian Virus 40 (SV40), and human cytomegalovirus. DNA sequences derived from the SV40 viral genome, for example, SV40 origin, early and late promoter, enhancer, splice, and polyadenylation sites may be used to provide other gene~c elements for expression of a structural gene sequence in a malslrnalian host cell. Viral early and late 5 promoters are particularly useful because both are easily obtained from a viral genome as a fragment which may also contain a viral origin of replication (Fiers et al., Nature 273:113, 1978). Smaller or larger SV40 fragments may also be used, provided the approximately 250 bp sequence extending from the Hind III site toward the Bgl I site located in the SV40 viral origin of replication site is included.
Exemplary expression vectors for use in mammalian host cells can be constructed as disclosed by Okayama and Berg (Mol. Cell Biol. 3:280, 1983). A
useful system for stable high level expression of mammalian cDNAs in C127 murinemammary epithelial cells can be constructed subslantially as described by Cosman et al.
IMol. immunol. ~3:93S, 1986). A useful high expression vector, PMLSV Nl/N4 ~5 described by Cosman et al., Na~ure 31Z:768, 1984 has been deposited as ATCC
39890. Additional useful mamrnalian expression vectors are described in EP-A-0367566, and in U.S. Patent Application Serial No. 07n01,415, filed May 16 1991,incorporated by reference herein. The vectors may be denved from retroviruses. To achieve secretion of CD30 (a type 11 pro~ein lacking a native signal sequence), a 20 heterologous signal sequence may be added. Examples of signal peptides useful in mammalian expression sys~ems are the signal sequence for interleukin-7 (lL-7) descnbed in United States Patent 4,965~195; the signal sequence for interleukin-~receptor described in Cosm~n et al., Nature 312 :768 ( 1984); the interleukin-4 signal peptide described in EP 367.566; the type I interleukin- I receptor signal peptide ~5 des~ribed in U.S. Patent 4.968,607? and the ~ype II interleukin- I receptor signal peptide described in EP 460,846. Each of these references descnbing signal peptides is hereby incorporated by reference.
The present invention provides substantially homogeneous CD30-1, protein, ~ -which may be produced by recombinant exp~ession syslems as described ab~ve or 30 purified from naturally occurring cells. The CD30-L is puri~led to substantial homogeneity, as indicated by a single protein band upon analysis by SDS-polyacrylamide gel elec~ophoresis (SI)S-PAGE~).
In one embodiment of the present invention, CD3~1, is puIified fr~m a cellular source using any suitable p~tein purificanon techni~que. The cells may, for example, 35 be ac~ivated T-lymphocytes from a mammalian species of interest, such as the munne cell line 7B9 described in examples 2 and 3 ~r induced human peripheral blood T-cells.
An alternative process ~or producing the CD30-L protein cornprises culturing a host cell ~ansfo~ned with an expression vector compnsing a DNA sequence that WO 93/2~135 PCI /USg3/&4926 '2.~3~0~3 18 encodes CD30-L under conditions such that CD30-L is expressecl. The CD30-L
protein is then recovered from culture medium or ce}l extracts, depending upon the expression system employed. As the skilled ar~san will recognize, procedures forpurifying the rec~mbinant CD30-L will vary according to such factors as the type of S host cells employed and whether or not the CD-30-L is secreted into the culture medium.
For example, when expression systems that secrete the recombinant protein are employed, the culture medium first may be concentrated using a commercially available protein concentration filter, for example, an Amicon or Millipore Pellicon ultrafiltration 10 unit. Fnllowing the concentra~ion step, the concentrate can be applied ~o a purification matrix such as a gel filtra~ion medium. Alternatively, an anion exchange resin can be employed, for example, a matrix or substrate having pendant diethylar;ninoethyl ~DEAE) groups. The matrices can be acrylamide, agarose, dextran, cellulose or other types comrnonly employed in protein purification. Alternatively, a cation exchange step ~,~
15 can be employed. Suitable cation exchangers include various insoluble matlices comprising sulfopropyl or carboxymethyl groups. Sulfopropyl groups are preferred.
Finally, one or more reversed-phase high performance liquid ehron;latography (RP-HPLC) steps employing hydrophobic RP-HPLC rnedia, (e.g., silica gel having pendant methyl or other a!ipha~ic groups) can be employed to further puri~y CD3~L.
20 Some or all of the foregoing purification steps, in various combinations, can be employed to provide a substan~ially homogeneous recombinant protein.
It is also possible to utilize an affinity column comprising the ligand binding domain of CD30 to affinity-purify expressed CD30-1, polypeptides. CD30-L ~ ~;
polypeptides can be removed from an affinity column in a high salt elution buffer and ~5 then dialyzed into a lower salt buffer for use. Alternatively, the affinity column may comprise an antibody that binds CD30-L. Example S describes a procedure for employing the CD3~L protein of the present invention to genera~e monoclonal antibodies directed against CD30-L.
Recombinant protein produced in bacterial culture is usually isolated by initial30 disruption of the host cells, centnfugation, extraction from cell pellets if an insoluble polypeptide, or from the supernatant fluid if a soluble polypeptide, followed by one or more concentration7 salting-out, ion exchange, af~mity purification or size exclusion ch~omatography steps. Finally, RP-HPLC can be employed ~or final p~i~cation steps. Microbial ee!ls can be dismpted by any convenient method, including freeze-35 thaw eycling, sonication, mechanical disruption, or use of cell Iysing agents.
Transfolmed yeast host cells are preferably employed to express CD30-L as a secreted polypeptide. This simplifies purification. Secreted recombinant polypeptide from a yeast host cell fe~nentation can be purified by methods analogous to those WO 93124135 PCI`/US93/~4926 ., 19 ~¦ ~J 80~
disclosed by Urdal et al. (J. Chromatog. 296:171, 1984). rd et al. describe tWO
sequen~ial, reversed-phase HPLC steps for purification of recombinant human IL-2 on a preparative HPLC column.
The present imention further provides antisense or sense oligonucleotides comprising a single-stranded nucleic acid sequence (either RNA or DNA) capable of binding to target CD3~L mRNA (sense) or CD30-L DNA (antisense) sequences.
Antisense or sense oligonucleotides, according to the present invention, comprise a fragment of the coding region of CD30-L cDNA. Such a fragment gene~lly comprisesat least about 14 nucleotides, preferably from about 14 to about 30 nucleotides. The ability to create an antisense or a sense oligonucleotide, based upon a cI~NA sequence for a given protein is described in, for example, Stein and Cohen, Cancer Res.
48:~659, 1988 and van der Krol et al., BioTechniques 6:958, 1988.
Binding of antisense or sense oligonucleotides to target nucleic acid sequences results in the formation of duplexes that block translation (RNA) or r;anscription (DNA) by one of several means, including~enhanced degradation of the duplexes, premature termination of transcription or translation, or by other means. The antisense oligonucleotides thus may be used to block expression of CD30-L proteins.
Antisense or sense oligonucleotides funher compnse oligonucleotides having modified sugar-phosphodiester backbones ~or other sugar linkages, such as those described in WO91/06629)~ and~wherein such sugar linkages are resistant to endogenous m~cleases. Such oligonuc!eotides with resistant sugar linkages are stable in ViVQ ~i.e., capable of resisting enzymatic ~degradation) but retain sequence specificity to be able ~o bind to target nucleotide sequences. Other examples of sense or antisense -o ligonucleotides include those oligonucleo~ides which are covalently linked to organic 5 ~ ~ moleties, ;such as those described in WO 90J10448, and other moieties that increases àffinlty of ~the oilgonucleotide for ~a target ~nucleic~ acid ~sequence, suGh as poly-(L-Iysine). ~urther still, intercalaung agents, such as ellipticine, and alkylating agents or metal complexes may~ attached to sense or antisense oligonucleotides to modify binding specificities of the an~isense or sense oliginucleotide for ~he target nucleotide sequence. Antisense or sense oligonocleotides may be intr~duced into a cell containing the target nucleic acid sequencelby any gene transfer method, including, for example, CaPO4-mediated DNA transfection, eleG~roporation, or other gene trans~er vectors such as Epstein-Barr virus. Aneisense or sense~oiigonucleotides are preferably introduced into a cell containing the ~arget nucleic acid sequence by inser~ion of the antisense or 35~ sense oligonucleotide into a sultable;ret~oviral vector, then contacting ~e cell with ~he re~ovirus vector containing the inserted sequence, either in YiVo or ex vivo. Suitable retroviral vectors include,~but are not ilmited to, those denved from the murine : ~

WO 93/24135 . PCr/USg3/04926 retrovih -MuLV, N2 ~a re :rovirus derived from M-MuLV), or or the double copy vectors designated DCI~SA, DCT5B and DCT5C (see PCI` Application US 91)/02656).
Sense or an~sense oligonucleotides may also be introduced into a cell containing the target nucleotide sequence by formation of a conjugate with a ligand S binding molecule, as described in WO 91/~4753. Suitable ligand binding molecules include~ but are not linuted to, cell surface receptors, growth facts:)rs, other cytokines, or other ligands tbat bind to cell su~rface recepto~s. Preferably, conjugation of the `-`
ligand binding molecule does not substan~ially interfere with the ability of the ligand binding molecule to bind to its corresponding molecule or receptor, or block entry of 10 the sense or antisense oligonucleotide or its conjugated version into ~he cell.
Alternatively, a sense or an antisense oligonucleotide may be in~oduced into a cell containing the target nucleic acid sequence by fonnation of an oligonucleotide-lipid complex, as described in WO 90/10448. The sense or antisense oligonucleotide-lipid complex is preferably dissociated within the cell by an éndogenous lipase.
The following examples are provided to illustrate particular embodiments and not to limit the scope of the invention.

l`his example describes construc~ion of a CD30/Fc-encoding vector to express a 20 soluble CD30/Fc fusion protein for use in detec~ing cDNA clones encoding a CD30 gand. A cDNA fragment encoding the extracellular region ~ligand binding domain) of the CD30 human receptor was obtained using polymerase chain reaction (PCR) techniques, and is based~ upon the sequence published by Durkop et al. ~Cell 68:421, 1992) and presented herein in Figure 1.
~5 The CD30 cDNA used as a template in the PCR reaction was prepared as follows.~ Total RNA was~ Isolated from a virally ~ansforrned human T-cell line designated HUT 102E. This cell line was derived by transfom~ing T-cells with human T-cell lymphotropic vi~us l~(HTLV-l) as described by Poiesz et al. (PNAS USA
77:7415-19, 1980). hrst strand c~NA was prepared using a SuperScrip~TM cDNA
synthesis kit available from (iIBC(~ RL (Gaithersburg, Maryland). The resulting single-s~anded cDNA was emplc>yed as the template in a PCR reaction.
The ~' primer employed in ~he PCR reaction was a single-stranded oligonucleo~ide (34-mer) of the sequence:
5' ATA5j~9g~CACCATGCGCGTCCTCCTCGCCGCGCI~G 3' This primer comprises a recognition site for the restriction endonuclease NotI
(underlined) upstream of a sequence (double under~ine) encoding the first (N-terminal) eight amino acids of the CD30 sequence shown in Figure 1~ from methionine (encoded by the translation initiation codon ATG) through leucine at position eight.

WO 93/24135 PClI /USg3/0~926 The 3' prirner employed in the PCR reac~ion was a single-stranded oligonucleo~ide (39-mer) of the sequence:
3' CAGCGAG~GAGGAGGTGCCCCl~CCrCGGGTCTAGAACA S' -This primer comprises a sequence (double underline~ that is complementaIy to S the sequence that encodes the last eight amino acids of the CD30 ex~acellular domain, i.e., amino acids 372 (Val) through 379 (Lys~ shown in Figure 1. The sequence CTCGGG that follows the CD30 sequence is complementary to codons for Glu and Pro. Glu and Pro are the first two amino acids of an antibody Fc fragment that is fused to the C-terminus of the CD30 fragment as described below. The pnmer also positions 10 a recognition site for the res~iction endonuclease Bgl~I (underlined) downstream. ~or use in attaching a DNA sequence encoding the remainder of the Fc-encoding gene.
The PCR reaction may be conducted using any suitable procedure, such as those described in Sarki et al., Science 239:487 ( 1988); in Recombinant DNA
Methodolog~, Wu et al.. eds., Academic Press Inc., San Diego ( 19~9), pp. 189-196;
is and in PCR Protocols: A Cuide to Methoc~s and Applic~tions, lnnis et al., eds., Academic Press, lnc. (1990). An example of a suitable PCR proeedure is as follows.
All tempera~ures are in degrees centigrade. The following PCR reagents are added tO a 0.5 ml Eppendorf mierofuge tube: 10 ~11 of lOX PCR buffer (S00 mM KCI, 100 mM
Tris-HCI7 ~pH 8.3 at 25C, 25 mM MgCI~, and 1 mglml gelatin) (Perkins-Elmer Cetus, 20 ~ Non~alk, CN~, 8 111 of a 2.5 mM solution containing each dNTP (2 IT~I dATP, 2mM
dCI'P? 2mM dGTP and 2 mM dl~P), 2.5 units (0.5 ~1 of standard 5000 units/ml soIution) of Taq DNA polymerase (Perlcins-Flmer Cetus), 1 ng of te~nplate DNA, lO0 piGomoles of each of the oligonucleotide pnmers, an~ water to a fina~ volume of 100 ~.
The ~lnal mixture is then overlaid with lOO ~1 parafin oil. PCR is carried out using a '~5 DN~ thennal cycler ~Encomp9 Sall Diego, ~A).
In a pre~erred procedure, the template was denatured at 94 for S minutes, ~; ~ollowed by 5 cycles of 94 for 1 minute (denatura~ion), 48 for 1 min. (annealing), and 72" for 1 min. (extension); ~ollowed by 30 cycles of 94 for 1 min., ~8- for 1 min., and 72- for 1 min.. with the las~ cycle b~ing followed by a final extension at 72- for 5 mins.
30 An aliquot of the products of this PCR reaction was reamplified in a second PCR
reactios~, using the same conditions.
The desired I)N~ fragrnent amplified by this PCR reaction comprised a NotI
site upstream of a sequence encoding the entire extracellular domain of CD30. followed by a BglII site. The PCR reaction products were digested wi~h Nod and BglII, a~nd the 35 desired fragment was punfied by gel electrophoresis.
A DNA sequence ~encoding an antibody Fc fragmen~, to be fused ~o the CD30-encoding DNA fMgment, was prepared as follows. DNA encoding a single chain polypeptide derived from the Fc region of a human IgG 1 antibody has been cloned into the SpeI site of the pBLlJESCRlPT SK~ vector, which is available from S~atagene (~loning Systems, La 3011a, California. This plasmid vector is replicable in E. coli and contains a polylinker segment that includes 21 unique restriction sites. The DNA and encoded arnino acid sequences o~ the cloned Fc cDNA coding region are presented in S Figure 2. A unique BgllI site has been introduced near the 5' end of the inserted Fc encoding sequence as shown in Figure 2.
The Fc polypeptide en~ed by the DNA extends from the N-terminal hinge region to the native C-terminus, i.e., is an essentially ~ull-length antibody Fc region.
Fragments of Fc regions, e.g., those that are truncated at the C-terminal end, also may be employed. The fragments preferably contain multiple cysteine residues (at least the cysteine residues in the hinge reaction) to permit interchain disulfide bonds to forrn between the Fc polypeptide portions of ~wo separate CD30/Fc fusion proteins, fo~ning dimers as discussed above.
The recombinant vector containing the Fc sequence is digested with BglIl is (which cleaves only at the site shown in Figure 2) and Notl (which cleaves the vector in the multiple cloning site downstream of the Fc cDNA insert. The Fc-encoding fragment (about 720 bp in length) was isolated by conventional procedures using LMT
agarose gel elec~ophoresis.
The Notl/Bglll CD30-encoding DNA fragment and the Bg~ll/Notl Fc-encoding DNA fragment prepar~ above were ligated into an expression vector designated pDC406 as follows. Plasmid pl:3C406, which has been described by McMahan et al.
(E~B~ J. Iû:'~821, 19~1), is an expression vector for use in mammalian cells, but is also replicable in E. coli cells.
pDC406 contains origins of replica~on der~ved from SV40, Epstein-Barr virus and pBR32~ and is a derivative of HAV-E0 described by Dower et al., J. lmmunol.
142:4314 (1989~. pDC406 di~fers from HAV-E0 by tne deletion of the intron present in the adenovirus 2 tripartite leader sequence in HAV-ES:). pD~406 was digested with NotI, which cleaves the plasmid in a multiple cloning site ~ust 3' of the Sall site, then treated with calf intestine alkaline phosphatase (CIAP) ~o prevent self ligation.
A three-way ligation to join the vector, Fc~ and CD30 DNA ~ragments was conducced under conven~ional conditions, and Æ. coli cells were transformed with ~he ligation mixture. A plasmid of the desi~ed siæ that was recovered from ~he E. coli cells was ~ound to comprise the CD30/Fc gene fusion insert, but in the wrong orienta~ion for expression. The CD3~/Fc; gene fusion was excised from this recombinant plasmid by Nod digestion and ligated tO NotI-diges~ed and CIAP-~eated pDC406. E. cofi cellswere transfonned with the liga~on mixture. A recombinant plasmid containing the insert in the desi~ed orienta~ion was isola~ed. The CD30 sequence was fused (in the s~ne reading frame) to the downstream Fc sequence.

W~ g3/2413~ P~r/US93/o4926 ~3 CD30/E~c fusion molecules preferably are synthesized in recombinant mamrnalian cell culture because they are generally too large and complex tO be synthesized by prokaryotic expression methods. Examples of suitable mammalian cells for expressing a receptor/Fc fusion protein include CV- 1 cells (ATCC CCL 70) and S COS-7 cells (ATCC CRL 1651), both derived from monkey kidney.
The DNA construct pDC406/CD30/Fc was transfected into the monkey kidney cell line CV-1/EBNA (ATC~ CRL 10478). In mammalian host cells such as CV lIEBNA, the CD30/Fc fusion protein is expressed off the HIV transactivating region (TAR) promoter. The CV- 1/EBNA cell line was derived by transfec~on of the CV- 1 cell line (ATCC CCI_ 70) with a gene encoding Epstein-Barr virus nuclear antigen- 1 ~EBNA- 1 ) that constitutively expresses EBNA- 1 driven from the human CMV intermediate-early enhancer/promoter as described by McMahan et al., supra. : :
The EBNA- I gene allows for episomal replicauon of expression vectors, such as pDC4~)6, that contain the EBV ongin of replication.
i5 CVI-EBNA cells transfected with ~he pDC406/CD30/Fc vector were cultivated in roller bottles to allow transient expression of the fusion protein, which is secreted into the culture medium via the CD30 signal peptide. Tlle CD3~/Fc fusion protein was - purified by afflnity chromatvgraphy. Briefly, one liter of culture supernatant containing th~ CD30/Fc fusion protein was purified by ~lltering the supernatants (e.g., in a 0~45 filler) and applying the filtr~e to a protein G affinity column (Schleicher and Schuell, ; Keene, NH) according to manufacturer's inst~uctions. The Fc portion of the fusion protein is bound by the Protein G on the column. E~ound fusion protein was eluted from the column and the purity confirmed on a silver stained SDS gel.
:
1~ ~~~
This example describes screening of certain cell lines for the ability tO bind aCD30/l;c fuslon protein. T hose cell lines found to ~ capable of binding CD30/Fc were considered tO be candidates for use as nucleic acid sources in the attempt to clone (~D30-L.
~j nyla~ono CD~FcFusion ~roteins The purified C~D30/Fc fusion pr~tein prepared in Example 1 was labeled with bio~in for use in screening cell lines. CD3Q/Fc or control human IL-4R/Fc were bio~nylated as follows: 50 llg protein (200 500 ,uglml in O.lM NaHCO3 pH 8.3) was incubated with 2,ug ( 1 mg/ml in DMSO) Biotin-X-N-hydroxysuccinimide (NHS, Calbiochem, La Jolla, CA) for 30 min at room temperature. At ~he end of the incubation period, the reac~ion mixture was microfuged through a 1 ml Sephadex G-25 (Pharmacia) desalting column and the eluate adjusted to 100 ~lg/ml in PBS plus 0.02%
, .
'' WO 93/24135 PCI/US93/~4926 ~3~3 ~4 NaN3. Protein concen~ation of biotinylated CD30/Fc and hIL-4R/Fc was determined by n~icro-E~CA assay (Pierce, Rockford, IL) with ultrapure bovine serum albumin as standard.

S nOw cvtometric stainin~ with biotinvlat_d Fc fusion proteins Cell lines such as those iden~fied below are screened for binding of biotinylated CD30/Fc by the following procedure. Staining of 1x106 cells was carried out in round-bottomed 96-well microti~er plates in a volume of 20 ~ ells were pre-incubated for 30 min at 4C with 50 1l1 blocking solution consis~ng of 100 llg/rnl human IgGl + 2%
goat serum in PBS + azide to prevent non-specific binding of labeled fusion proteins to Fc receptors. 150 1l1 PBS + azide was then added to the wells and cells were pelleted by centrifugation for 4 min at 1200 rpm. Pellets were resuspended in 20 ~11 of 5 ,ug/ml biotinylated CD30/Fc or biotinylated hIL-4R/Fc (as a specificity control) diluted in blocking solution. After 30-45 min incubation at 4UC, cells were washed X2 in PBS +
azide and resuspended in 20 1l1 streptavidin-phycoerythrin (Becton Dickinson) diluted 1:5 in PBS + azide. After an additional 30 min, cells are washed x2 and are ready for analysis. If necessary, stained cells can be fixed in 1% formaldehyde, 1% ~etal bovine serum in PBS + azide and stored at 4C in the dark for analysis at a later time.Streptavidin binds to the biotin molecule which was attached to the CD30/Fc protein. Phycoerythrin is a fluorescent phycobiliprotein which serves as a detectable label. The level of fluorescence signal was ~hen measured for each cell type using a FA~Scan(~) ~ow cytometer ~Becton Dickinsvn).

Cell Ljnes to be Screened for CD30/Fc Bin~
'~5 Sheep red blood cell (SRBC)-specific helper T-cell lines designated 7~2 (TH1), 7B9 ~ and SBEI 1 (TH2) were derived by limiting dilution ~rom primary antigen-induced cultures of murine C57BL/6 spleen cells. TH phenotypes of these clones were determined by their ability to secr~te L-2 and/or L-A in response to stimulation with the mltogen concanavalln A ~(~onA~.
Human peripheral blood T-cells were stimula~d for 16 hours with 10 llg/ml of an anti- CD3 monoclonal antib~dy immobilized on plas~c, prior tO assay for CD30/Fc binding. The an~i-CD3 MAb stimulates~the T-cells through ehe CD3-T-cell receptor(TCE~) complex~

'~

WC~ 93/24135 PCI/US93/04926 2~ 3~0~
~5 Biotinylated CD30/Fc bindin~, Murine T-cell lines 7C2, 7B9 and SBEl I showed signi~lcant binding of biotinylated CD30/Fc over that seen with control lL-4R/Fe, after stimulation for 18 S hours wi~h 3 tlg/ml Con A. 7C2 cells were also assayed after 6 hours stimulation with Con A, and specific binding o~ labeled CD30/Fc was seen. The anti-CD3 MAb activated human T-cells showed significant hinding of biotinylated Cl:)30/~c. Binding of biotinylated CD30/Fc was not detected on any of these cell lines in the absence of stimulation.
Any of the cell lines that demonstrated binding of CD30/Fc may be used as a source of nucleic acid in an attempt to isolate a CD3~L-encoding ~)NA sequence. A
cDNA lib~ary may be prepared from any of the three Con A stimulated murine T-cell lines or the activated human peripheral blood T-cells, and screened to identity CD3~L
cDNA using the direct expression cloning strategy,described below, for example.
S Other types of activated T-cells may be screened for CD30 binding to identify additional sui~able nucleic acid sources. The cells may be derived from human, murine, or other mammalian sources, including but not limi~ed to rat, bovine, porcine, or va~iousprimate cells. ~ her, ~he T-cells may be stimulated with mitogens other than ConA or otherwise activated by conventional techniques. It is to be noted that human CD30/Fc was successfully employed to screen both human and murine cell lines in the foregoing assay (i.e., human CD30/Fc binds to a ligand on both the human and the murine cell , - lines tested).

~
This example describes p~epara~ion of a cDNA library for expression cloning of murine CD3~L. The libra~y was prepared from the m~ne helper T-cell line designated ~B9 (described above and in Mosley et al., Cell 59:335, 1989), which was s~imulated for 6 hours with 3 ~g/ml Con A. The libra~y construction technique was substanaally similar to that described by Ausubel et al., eds., Current Protocols In Molec~lar Biology, Vol . 1, ( 1987). Briefly, total RNA was ex~acted ~rom the 7B9 cell line and poly (A)+ mRNA was isolated by oligo dT eellulose chroma~ography. , ~, Double-stranded cD~lA was made substantially as described by Gubler et al., Gene25:263, 19$3. Poly(A)+ mRNA fragments were converted to RNA-cDNA hybrids by reve~se ~anscriptase using random hexanucleotides as prime~s. The RNA-cDNA
hybrids were then converted ineo double-soanded cI)NA fragments using RNAase H in combination with DNA polymerase I. The resul~ng double-s~anded cDNA was blunt-ended wi~ T4 DNA polymerase.

WO 93J~4135 PCI /US93tO~lg26 Unkinased (i.e. unphosphorylated3 BglIl adaptors:
5'- GATCTGGCAACGAAGGTACCATGG -3' ACCGTTGCTTCCATGGTACC -5' were ligated to S' ends of the resulting blunt-ended cDNA, using the adaptor cloning S metho~ described in Haymerle et al.7 Nucleic Acids Res. 14:8615, 1986. Only the 24-mer oligonucleotide (top strand~ will covalently bond to the cDNA during the ligation reaction. Non-covalently bound adaptors (including ~he 20-mer oligonucleotide above) were removed by gel fil~adon chromatography at 68C. This left 24 nucleotide non-s~lf-complemen~y overhangs on cDNA. The cDNA was inserted into pDC202, a mamrnalian expression vector that also replicates in E. coli. pDC202 is derived from pDC201 (Sims et al., Nature 241:585, 1988). The plasmid pCD201 was assembled from (i) the SV40 origin o replication, enhancer, and early and late promoters; (ii) the adenovirus-2 major late promoter and tripa~te leader, (iii) SV40 polyadenylation and transcnption termination signals; (lv) adenovirus-2 virus-associated RNA genes (VAI
is and VAII); and ~v) pMSLV (Cosman et al., Nature 312:768, 1984). The multiple eloning site contains recognition sites ~or Kpn I, Sma 1, and Bgl Il. Certain extraneous vector sequences bordenng the VA genes were excised from pDC201 to create pDC202. Each of the above-narned features of pDC201 is present in pDC202 as well.
pDC202 was digested with BgnI and Bgl Il adaptors were ligated thereto as 20 described for the cDNA above, except that the bottom strand of the adaptor (the 2~
mer) is covalently b~und to the veetor, rather than the 24-mer ligated to the cDNA. A
single-stranded extension complementary to that added to the cDNA thus was added tO
the Bglll-digested vec~or. The S' ends of ,he adaptored vector and cDNA were phosphorylated and the two DNA species were then ligated in the presence of T4 ?5 polynucleotide kinase.: Dialyzed ligation mixnlres were electroporated into E. col~ strain D}15a and transformants selected Oll ampicillin plates.
To create an expression cloning library, the recombinant vectors containing 7B9-derived cDNA were ~ns~erred from E. coli to mammalian host cells. Plasmid DNA was isolated from:pools:of ~ransformed E. coli and transfected intv a sub-30 confluen2 layer of COS-7 cells using standard teehniques. The ~ansfected cells were cul~ured ~or ~wo to three àays on charnbered glass slides (Lab-Tek) to permit transient expression of the inserted DNA sequences.

This example describes screening of the expression cloning library m~de in Example 3 with a labeled CD30iFc fusion protein. The purified CD30/Fc fusion protein prepared in Example I was radioiodina~ed with 125I using a commercially available solid phase agent (IODO-GEN, Pierce). In this procedure, S ~g of IOI)O-WO g3/2413~ Pcr/US93/04926 27 2i31~
GEN were plated at the bottom of a l O x 75 mm glass tube and incubated for twenty minutes at 4 C with 75 ill of 0.1 M sodium phosphate, pH 7.4 and 20 ~11 (2 mCi)Nal25i. The solution was then transferred to a second glass tube containing S ~lg of CD30/Fc in 45 ,ul PBS and this reaction mixture was incubated for twenty minutes at 4 5 C. The reaction mixture was fractionated by gel filtration on a 2 ml bed volume of Sephadex~) G-25 (Sigma), and then equilibrated in RPMI 1640 medium containing 2.5% (v/v) bovine serum albumin (BSA), 0.2% (v/v) sodium azide and 20 mM Hepes, pH 7.4 binding medium. The ~mal pool of 1251 CD30~Fc was diluted to a working stock solution of 1 x 10-7 M in binding medium, which may be stored for up to one 10 month at 4q C without detectable loss of receptor binding activity.
Monolayers of transfected COS-7 cells made in Example 3 were assayed by slide autoradiography ~or expression of CD30-L using the radioiodinated CD30/Fc fusion p~otein. The slide autoradio~raphic technique was essentially as described by Gearing et al., EMBO J. 8:3667, 1989. Briefly, transfected COS-7 cells were washed once with binding medium (RPMI 1640 containing 25 mg/ml bovine serum albumin ~BSA), 2 mg/ml sodiurn azide. 20 mM Hepes pH 7.2, and 50 m~lml nonfa~ dry milk) and incubated for 2 hours at 4C in binding medium containing 1 x lû-9 M 12~1-CD30/Fc fusion protein. After incubation, cells in the chambered slides were washed three times with binding buffer, followed by two washes with PBS, ~pH 7.3) to 20 remove unbound radiolabeled fusion protein.
The cells were fjxed by incubating in 10% gluteraldehyde in PBS ~30 minutes at room temperature), washed twice in PBS and air-dried. The slides were dipped in Kodak GTNB-2 photographic emulsion (5x dilution in water) and exposed in the dark for two to four days days at room temperature in a light-proof box. The slides were 5 develop~d in Kodak Dl9 developer, rinsed in water and ~lxed in Agfa G433C fixer.
The slides were individually examined under a microscope at 25-40x ma~nifica~on.Positive slides showing cells expressing CD30-L were identified by the presenee of autoradiographic silver ~rains against a light background.
Eight pocls, each containing approximately 2000 individual clones, were 30 identified as positive îor binding the CD3(t/Fc fusion protein. Two pools were titred and plated to provide plat~s containing approximately 200 colonies each. A replica of each bre~kdown pool was made and the cells were scraped to provide pooled plasmid DNA for transfection into COS-7 cells. The smaller pools were screened by slide autoradiography as described previously. Several of the breakdown pools contained 35 elones tha~ were posiive for CD30-~ as indicated by the presence of an expressed gene pro~uct capable of binding to the CD30/Fc fusion p~o~ein.
Individual colonies from two of the breakdown pools were picked ~om the replicas and inoculated into culture medium in individual wells of 9~well plates.

W~ 93/~q135 P~/US93/04926 ~3~3 ~8 Cultures were mixed by pooling rows and columns and the mixed cultures were used to prepare DNA for a final round of transfection and screening. An intersection of a positive row and and a positive column identified the positive colony. DNA from the pure clone was isolate~, retransfected and rescreened.
The recombinant plasmid cont?ining murine CD3~L cDlYA was recovered from the pure clone (COS-7 ho~t cells~ and transformed into E. coli strain DH5a. The mammalian expression vector pDC202 containing murine CD3~L cI~NA (designated pDC202-mCD30-L~ was deposited in E. coli strain DHSa host cells with the American l'ype Culture Collection, Rockville, MD (ATCC) on May 28, 1992, under accession number ATCC 69004. The deposit was made under the eenns of the Budapest Treaty.
A DNA sequence for the coding region of the cDNA insert s)f clone pOC202-mCD30-L is presented in Figure 3, along with the encoded amino acid sequence. The protein comprises an N-terrninal cytoplasrnic domain (amino acids 1-27), a transrnembrane region (arnino acids '~8-48), and an extracellular. i.e., receptor-binding is domain ~an~ino acids 49-2'70). This protein lacks a signal peptide.
Six amino acid triplets constituting N-linked glycosylation sites are found at amino acids 56-58, 67-69, 95-97, 139-141, 175-177, and 187-189 of ~Igure 3. The protein comprises no KEX2 protease processing sites.
In this particular vector construction, an AT~i codon located in the Bgl 11 adaptors (see Example 3) is in the same reading frame as the CD30-L cDNA insert.Thus, a percentage of the transcripts may comprise the following DNA sequence ups~eam of the sequence of Figure 3. The encoded amino acids are also shown. andwould be fused to the N-terminus o~ the Figure 3 sequence. but are not CD30-L-specific amino acids.
S ATG GGC TGT GGG GCI CCT TCC ccr GAC CCA GCC
Met Gly Cys Gly Ala Pro Ser Rro Asp Pro Ala This example illustrates the preparation of monoclonal annbodies to CD30-L.
CD30-L is expressed in mammalian host cells such as COS-7 or CV 1-EBNA cells andpurified using CD3Q/Fc af~lnity chromatography as described herein. Purified CD30- ~ -L can be used to generate monoclonal antibodies against CD30-L using conventional techniques, for example, those techniques described in U.S. Paeent 4,41 1 ,9g3. 1 he immunogen may compnse a protein (or fragment thereof, such as the ex~acellular domain) fused to the peptide Asp-Tyr-Lys-Asp-Asp-Asp-Asp-Lys (l:)YKDl:)DDK) (Hopp e~ al., BiolTechnology 6:1204, 1988 and U.S. Pa2ent No. S,011,912~ or fused to the Fc portlon of an antibody, as described above.
' ' WO 93/24135 PCI'/U~93/04926 ~3~Q~3 Briefly, mice are immunized with CD30-L as an immunogen emulsified in complete Freund's adjuvant, and injected in amounts ranging from 10~ g subcutaneously or intraperitoneally. Ten to ~welve days later, the immunized animals are boos~ed with addi!~ional CD30-L emulsified in incomplete Freund's adjuvant. Mice S are periodically bcosted thereafter on a weekly to bi-weekly immunization schedule.
Serum samples are periodically taken by retro-orbital bleeding or tail-tip excision for - testing by dot blot assay or ELISA (Enzyme-Linked Immunosorbent Assay~, for CD30-L antibodies.
Following detection of an appropriate antibody titer, positive animals are provided one last intravenous injection of CD30-L in saline. Three to ~our days later, the animals are sacri~lced, spleen cells harvested, and spleen cells are fused to a murine myeloma cell line ~e.g., NS 1 or Ag 8.653). The latter myeloma cell line is available frorn the American Type Culture Collection as P3x63Ag8.653 (ATCC CRL 1580).
Fusions generate hybridoma ce}ls, which are plated in multiple microtiter plates in a I S HAT (hypoxanthine, aminopterin and thymidine) selec~ive medium to inhibit proliferation of non-fused cells, myeloma hybrids, and spleen cell hybrids.
The hybridoma cells are screened by ELISA for reactivity against pulified CD30-L by adaptations of the techniques disclosed in Engvall et al., I~nunochem.8:871, 1971 and in U.S. Patent 4,703,004. A preferred screening technique is theantibody capture technique described in l~eckmann et al., (J. Immunol. 144:4212,1990). Positive hybridoma cells can be injected intraperitoneally into syngel~eic - E~ALB/c mice IO produce~ascites containing high concentrations of anti-CD3~L
monoclonal antibodies. Alternatively, hybridoma cells can be grown in vitro in flasks or roller bottles ~y various techniques. Monoclonal an~ibodies produced in mouse ~-~
'~5 ascites can be pùrified by ~ammonium sulfate precipitation, ~llowed by gel exclusion chromatography. Altnatively,affinitychromatographybaseduponbindingof antibody to protein A or protein G can also be used, as can affinity chromatography ~;
based upon binding to CD30-L. -~

~
This exarnple illustrates a cross-species hybridiza~r>n technique which was usedto isolate a human CD30-L cDNA using a probe derived from the sequence of murine ~ -~
CD30-L. A murine CD30-L probe was produced by excising the entire cDNA inser~
from murine clone pDC202-mCD30-L (ATCC 69~04, described in Example 4) by Bgl II digesnon~ and 32P-labeling the fragment using random primers ~Boehringer-Mannheim).
A human peripheral blood Iymphocyte (PBL) cDNA libr~y was constructed in a phage vector (~gt 10). The PBL cells were obtained from noTmal human volunteers WO 93/24135 Pcr/usg3/o4926 ~,~3~Q~ 30 and trea~ed with 10 ng/ml of OKT3 (an anti-CD3 antibody), and 10 ng/ml of human Il,-2 (Irnmune~, Seattle, WA) for six days. The PBL cells were washed and stimulatedwith 500 ng/rn~ ionomycin ~Calbiochem) and 10 ng/rnl PMA (Sigrna) for four hours.
Messenger RNA was isolated from the stimulated PBL cells. cDNA synthesized on the mRNA template was paekaged into Agt 10 phage vec~ors (Gigapak(~ Stratagene, San Diego, CA) according to manufacturer's instructions. Recombinant phage were ~en plated on E. coli strain KW251^ and screened using standard plaque hybridizationtechniques.
The murine probe was hyb~idized to phage GDNA in the following hybridization buffer at 37 C overnight:

50% Formamide 20 mM Pipes (pH 6.4) 0.B M NaCl ~ mM EDTA
0.5% SDS
0~1 mg/ml salmon spermDNA
~Hybridization was followed by washing with 2X SSC, 0.1% SDS at ~0 C.
, Posi~ive ~hybridizing) plaques were visualized by autoradiography.
Six of the positive plaques were purified and tbe inserts were isolated by PCR
amplification using oligonucleotides that flank the cloning site. A partial amino.acid sequence for human CD30-L was derived by determining the nucleotide sequence of a portion~of one of these lnserts (clone #9, about 2.0 kb in length). This partial amino -; ~ acid sequencè is presented and aligned with the corresponding portion of muriné CD30-5~ ~ ~; L ln F~gure 4. The human sequence is in the top rows, indicated by (h), and the murine sequence is indicated by ~(m), with amino acids of uncertain identity being represented as X. The transm~mbrane region is underlined for the mouse sequence and overlined for the human sequence. ~
The first X ~at p~sition 6) in the human sequence is most likely a methionine 30 residue encoded by an initiation codon. As can be seen by reference to the murine sequence of Figure 3, an N-terminal fragment (amino acids 1-130) of murine CD3~Lis aligned with the par~ial human sequence in Figure 4.
The~DNA sequence of the en~ire coding region of the human ~D3~L clone was ùetemnined and is presented in Figure~5, along with the encoded amino acid sequence.
35 The N-te~minal cytophsmic domain tamino acids 1 to 21) is followed by a transsnembrane region (amino acids 22~ to 43, underlined in Figure 5) which is followed by the extracellular, i.e.,~receptor-binding domain (amino acids 44-215). This protein lacks a signal peptide. Where the partial human CD30-L of Figure 4 differs from the WO 93/24135 PCr/US93/0~92~

full length human sequence presented in Figure 5, the Figure S sequence is considered t~ be accurate. Comparison of the murine (Figure 3) and human (~igure 5) CD3~L
amino acid sequences using the above-described GAP comput~r program reveals 73%
identity and 83% similarity between the two sequences.
S Amino acid triplets that eons~itute potential N-linked glycosylation sites are found at positions 62-64, 90-92, 134-136, 170-172, and 182-184. A KE7C2 proteaseprocessing site is found at amino acids 72-73 of Figure 5. If desired, ~hese N-glycosylation prGcessing sites may be inactivated to preclude glycosyla~ion as described above. The KEX2 sites may be inactivated to reduce proteolysis when the CD30-L
protein is expressed in yeast cells, as described above.
The products of the above-described P~ reaction (by which the cDNA insert of the positive clone was amplified) we-re digested with EcoRI and ligated into an EcolV-digested vector designated pGEM~L. Plasmid pGEMBL is a derivative of the standard cloning vector pBR322 and conlains a polylinker having a unique EcoRI site along with several other unique restriction sites. The plasmid also comprises anampicillin resistance gene. An exemplary vector of this ~ype is described by Dente e~
al., (Nucl. Acids Res. 1 1:1645, 1983).
E. coli s~rain DH50~ was transfonned with the ~igation mixture and ~ansformants containing tbe desired recombinant plasmid were identified. Samples of E. coli DHSa containing plasmid hCD30-L/pGEMBL were deposited with the American Type Culture Collection, Rock~ille, MD (ATCC) on ~une 24, 1992, under accession number ATCC 69020. The deposit was made under the tenns of the Budapest Treaty. The deposi~ed recombinant plasmid con~ins human CD30-L DNA
that includes the complete coding region shown in Figure 5.
~5 :
EXAIVIPLE 7: Isolation of Murine and Human CD30~L DNA Encoding Because the CD30^L clones isola~ed in examp}es 4 and 6 had relatively short 5' noncoding regions and lacked stop codons upstream of the firs~ initiation codon~isola~ion of CD30-L DNA compnsing additional 5' s~quences was attempted. An anchored PCR technique was employed, generally as desc ribed by Loh et al., Science 243:217 (1989) and Ca~ier et al., Gene 116:173 (1992), both of which are hereby ~incorporated by reference. The same procedures were employed for isolaeing murine and human clones, except as noted.
3 5 Flrst strand cDNA was synthesized using a Superscnpt~ cDNA kit (GIBCO/BRL, Gaithersburg, MD~ on the following mRNA templates:
murine: 5~g total RNA from 7B9 cell line descn~ed in Example 3.
human: 21~lg poly A+ RNA from human peripheral bl~od T-cells (~he stimulated PBLs described in Example 6) .

WO 93/2413r. PCr/lJS93/04926 The primers employed in the cDNA synthesis (refeITed to as primers #l hereillafter) were:
murine: 5' ACiATGC~(~ACAC,~G 3' human: 5'ATÇ~ACCAGAl'rCCCATC 3' Murine pIimer #1 is complementary to nucleotides 265-281 of Figure 3.
Human primer #1 is complementary to nucleotides 325-341 of Figure 5.
The reaction rnixture was treated with RNAse H, then pUlified over a Sephadex G50 spin column (Sigma). After drying, the cDNA was resuspended in: 10 I,ll H20, 4 10 111 SX terminal deoxynucleotidyl transferase (TdT) buffer (as specified by GIBCO/BRL, Gaithersburg, MD~, 4 ,ul lmM dATP, and 1 ~11 TdT (15 units/~Ll). Thisreaction rr~ixture was incubated at 37C for 10 rninutes to add a poly-A tail ~o the 3' end of the cDNA. The reaction was stopped by heating at 68~C for 15 minutes, and themixture was applied to a Sephadex G50 spin column. The eluate was diluted to ~50 ~l :
with 10 rnM Tris ~pH 7.5), I mM EDTA. A first PCR reaction mixture was ~
prepared by combining: ;;
10 ,ul first strand cDNA ~tailed with adenines) :~
10 ~ll 10 X kuffer !1l 1st anchoring primer: 5' GCATGCGC9Ç~5l~GGAG(~TI7 3' ~0 (100 ng/~
,ul ~nd anchoring primer: S' GCAT(:;CGCGCGGCCGCGGAGGl-r 3' (lOC) ng/~) 2 ~Ll primer#2 (antisense) munne: S' ACAGAAGAGATCCICTG 3' ~5 ~ human: 5' CCAACACCATAATAGTG 3' Taq: DhlA polymerase O.~ 5 mM dNTP's Z3.2 ul dH~C) 100.0 ~ TAL
:30 The ~ollowing reaction condi~ions (temperature cycles3 were employed for this :
first PCR, and each of the PCRs described below:
94~C - 5 minutes - lX ::
94C - O.5minutes ~ -55C 1.5minutes 1 - 30X
72C - ~.~ minutes I
72 C - 5minu~es - 1X
The first anchoring primer contains a poly T segment tha~ will anneal to ~he poly A tail added to the cDNA. This p~imer also inserts a Notl restriction site (underlined) i nto the ampli~led DNA. The second anchonng primer anneals (in later cycles of the 40 reac~on) ~o Ihe NotI slte-containing sequence inserted into the amplified DMA via the first anchoring primer.

WO 93/~4135 1'~(~/US93/04926 3 L t~ ~ ~3 The murine primer #2 is complemen~y to nucleotides 20~222 of figure 3.
The human pnmer #2 is complementary to nucleotides 108-124 of figure S.
A second PCR reac~on n~ixture was prepared by combining:
25 ~11 first PCR reaction mixture (after the above reacnon) 2 ~1 2nd anchonng primer 2 ~1 primer #2 10 ',11 lOx buffer 0.8 ~1 25 mM dNl'Ps ,ul Taq DNA polymerase 1059.2 lLI dH20 :

A third PCR reac~ion mixture was prepared by combining: .
10~11 . nd PCR reac~ion mixture (a~ter completion of the reaction) 10',11 IOx buffer 2111 2nd anchoring primer 1 primer #3 murine: S' GGGTGACACrrGTG~CTCCAGGG 3' human: S' GGGT(~GACCAAGGCACAGAGCCA 3' 111 Taq DNA polymerase 0.8 ~1 25 mM dNTPs 7 ~ 1 dH20 10~). ~1 TOTAL -:' The murine primer ~3 contains a segment complementary to nucleotides 49-66 of figur~ 3. Human pnmer #3 contains a segment complementary to nucleotides 80-94 of figure 5. Each primer #3 also contains a segment that introduces a Sal~ res~iction site (underlined) into the amplified DNA. ~:
PCR reaction products ~from PCR reaction no. 2 for human and no. 3 for murine) were separated by elec~ophoresis vn a 1% MuSieve agarose gel (FMC ~:
Bioproducts~ Rockland, ~). A PCR band compnsing DNA of about 300 bp was isolated for both murine and human. The CD30 L DNA was further ampli~led in :-another PCR reac~ion. The reaction mixture comprised:
S ~I band fr~m gel ~melted at 68C) :::
10 111 lOx buffer :~
2nd anchoring primer 2 ~1 primer~3 -:
~I Taq DNA polymerase 0.8 ~11 25 mM dNTP's ~1 dH~O
100.0 111 TaI'~L

The nucleotide sequence of the reaction p~oducts was determined. The reaction products may be sequençed di~cely or subcloned by digesting with NotIJSall prior to 45 sequen~ng. Sequencing revealed additional DNA at the 5' end, compared to the clon~s of examples 4 and 6, including DNA encoding an addi~onal 19 N-terrrunal amino acids for both murine and hurnan CD30-L. DNA and encoded amino acid sequences ~or the WO 93/2413S P~/V~93/04g26 coding region of CD3~L DNA comprising this additional 5' coding sequence are shvwn in figures 6 (murine) and 7 (human). The additional N-tenninal amino acidscomprise no N-glycosylation or KEX2 protease processing sites.
The m~ine and human CD30-L DNAs isolated in this example were expressed S in CV 1 -EBNA cells. The molecular weight of the expressed protein, analyzed by non-reducing SDS-PAGE, was about 26,5~ daltons for murine and 26iO17 daltons for hurnan CD30-L. ~
Although the murine and human CD30-L proteins encoded by the clones of examples 4 and 6, respectively, are truncated at the N-terminus, the enc~ded proteins 10 are biologically active in that they bind to CD30. Thus, CD3~L proteins lacking from one to all of the first 19 amino acids shown in figures 6 or 7 are biologica~ly active CD30-L proteins of the present invention. Deletion of the first 19 amino acids of ; ~ figures 6 and 7 yields an amino acid sequence idenical to that presented in figures 3 and S, respectively.

~XAMPLE 8: An~ly~ of Biolo~i~al ACtiYi~i~S of C~
Cells on which CD30 expression has been previously observed were screened for a response to the recombinant CD30 ligand. The human eell.types scre~ned included activated T cells,~ three Hodgkin's lymphoma lines resembling H-RS cells with ; 20 ~ primitive B or T cell-like phenotypes, and a non-Hodgkin's lymphoma line of the large cell anaplastic Iymphoma (LCAL) type.
Pelipheral blood T-lymphocyte (PBT) cells were isolated by centrifugation over Histopague ~Sigma Chemical Co., St. Louis, MO) and ro~etting with 2-aminoethylisothiouronlum~bromide (AET)-treated sheep erythrocytes as described 5~ (ArmItage et al., Int. Immunol. ~:1039 (~1990)). The purified PBT were then cultured for 5~days in the presence of irnmobilized CD3 antibody and a titration of fixedCV1/EBNA cells expressing full length (membrane-bound) recombinant human CD30 gand. In contrast to con~ol cells transf~ted wlth vector alone, cells expressingCD3~L induced prolife~on of the stimulated T cells in a dose-dependent manner, with a maximal response observed with 2.5 x 104 (:~Vl/E~BNA cells/well. This enhanced proliferation (and othcr activities described below) could be blQcked by the nclusion of 1011g/ml~ of soluble CD30/Fc. A similar ability to induce proli~eration of CD3-activated T cells was~seen~in the presence of immobilized anti-CD30 monoclonal ~ antibody~ M44, suggesting the bivalen$ antibody ITnmics ligand-induced receptor cross 3~ linking. The~M44 rnonoclonal antiWdy is a mouse IgGl generated with purified CD30-Fc as imrnunogen. No response was seen to CD3~L in the absence Qf CD3 co-stimulaion.

WO 9~/2413S PCl /US93/04926 21~1Q~

The biological activity of CD30-L on human Iymphoma cell lines known tO
express CD30 was investigated. The CD30+ human lymphoma lines tested included HDLM-2, KM-H2, L-428, and Karpas 299 cells. Culture conditions for these four cell lines are published (Drexler et al., Leuk. Res. 10:487 ( 1986); Gruss et al., Cancer Res. ~ :~
5 52:33S3 (1992)).
The HD-derived cell line HDLM-2 was established from a malignant pleural effusion of a 74-year-old male with endstage IVB HD ~Drexler et al., 1986, supra;
Gruss et al., 1992, supra)~ HDLM-2 is phenotypically T-cell-like (Gruss et al., 1992, supra). KM-H2 and L-42B are B cell-like, HD-derived Iymphoma lines. The human Karpas 299 cell line was established from blast cells in the penpheral blood of a 25-year-old white male with the diagnosis of a large cell anaplastic lymphoma (Ki- I
posi~ive high-grade human Iymphoma)~ The peripheral blast cells with pleomorphicnuclei rcsembled plimitive histiocytes, which bear the surface markers CD4, CDS,HLA-DR and CD30, The Karpas 299 cell line possesses the same cytochemical.
immunologic, and chromosomal profile with a 2;5 translocation as the original peripheral blood blast cells of ~he patient (Fischer et al., Blood 72:234 (1988)).
The addition of CVl/EBNA cells (10,000 cells/well) expressing recombinant human CD30-L to the HD-derived cell line HDLM-2 (S0,000 cells/well~ resulted in enhanced proliferation, whereas addition of control CVI/EBNA eells transfected wi~h vector alone had minimal effect. The CD30-L-induced simula~on of HDLM-2 cell proliferation was time-dependent, with a maximal 3-4-fold enhancement observed at 72 hours. Similar results were obta;ned using immobilized M44 antibody, and the effect was dose-dependent. Cells cultured with an isotype-matched con~rol monoclonal anibody showed no response. Maxirnal enhancement of proliferation, a five-fold 5 increase over con~ol cultures, was detected after stimulation wi~h 10 ~g/ml of M44 for 72 hours. Here again, the M44 CD30 monoclonal antibody has agonist characteris~ics and~mimics pr~perties of the ligand. In con~ast to the above results, we could detect no CD30-L effects on proliferation or viability of the KM-H2 or L-428 cells, even though both lines were confiImed to be CD30+ by flow cytometry with M44.
A clear and dramatically different response to CD3~L was seen with the CD30+ non-Hodglcin Iymphoma ~LCAL) line Ka~pas 299. The addi~ion of either CVl/EBNA cells expressmg the CD30~L or M44 an~ibody to Karpas 299 cells (S x 103cells~wellj decreased the prolifera~ion eight-fold. This effect was further analyzed with cytotoxic assays measuring 5ICr-release. Both CVlIEBNA cells expressing CD3~L
and M44 anti~y induced specific 51Crrelease ~rom d~ese cells in a ~m~ and dose-dependent nsanner. At 18 hours, the specific release in response to CD30-L or M44 was 29.4% and 30.8%, respectively. The addition of :Vl/EBNA cells transfected with vec~or alone, or of an isotype-matched control an~bady, had no effec~. Thus, in , WO g3/24135 P~r/US93/~926 2~ ~Q~3 36 contrast to the enhanced proliferative response of the Hodgkin's Iymphoma-derived HDLM-2, the response of the Karpas 299 non-Hodgkin's Iymphoma line to CD30-L is cell death.

S ~L~ "'~

Various types of cells were analyzed by Northern blotting to detect CD30-L ;
transcripts (mRNA).
Human cells Human PBT cells, induced with a calcium ionophore, uninduced tonsillar T
cells and LPS-induced monocytes all expressed a single hybridizing transcript migrating between 18 and 28 S nbosomal RNA. IL-7-treated PBT cells, PMA treated tonsillar B cells, uninduced Jurkat or LPS activated THP-l macrophage, and GM-CSF
treated monocytes did not express CD30-L. L- l ,B induced low levels of CD3~L ini ~ monocytes. In addition, placefft-al tissue, the promyelocytic HL60 line and two Burkitt's Iymphoma B cell lines ~Daudi and Raji) were also negative for expression of D30-L transcripts. Thus human CD30-L expression was detected on specifically induced T célls and monocytes/macrophages.
.
Murine ce~
~20 ~ These results are mirrored in the murine system. LPS stimulated bone marrow-derived macrophage, Con A activated 7F9 T cells (similar to the 7B9 murine helper T-cell line described in examples 2 and 3) and an LPS stimulated subclone of the munne thymoma EL4 (EL:4 6.13 all~ express a single CD30-L transcript. Unstimulated EL4 6.1 and-7F9 cells, a bone marrow-derived stromal line D11 and a thyrnic stromal line F4, 25 ~ donotexpressCD30-L.~

Biochemir~al characterisncs of the recombinant, full-length cell surface fonns of munne and human CD30-L~ were assessed by surface radioiodina~ng cells transiently ~ -30 expressing the recombinant ligands, then isnmunoprecipita~ing the ligands with CD30/Fc (and protein G) ~m Iysates of detergent solubilized cells. Iodoacetarnide (20tnM) was included in Iysing and immunc~precipitation buffers to inhibit potential disulfide interchange. Washed precipitates were then displayed by SDS-PAGE with phosphorimaging. Colls transfected with Yeetor only~ or cells expressing recombinant 35 ~ ligand but immunoprecipitatcd with an isotype matched control ~huIgGl), showed no bands. Under reducin~ conditions, the dominant product for both hurnan and murine recombinant CD3()-L is a dîffwse 40 kd band. As the CD3~-L protein molecular weight is 26,~0 Kd, extensive use of the multiple N-linked~ glycosylation sites in the ~, WO ~3/2413~ Pcr/U~93/o49~6 2~ ~ 0~3 extracellular domains seems clear. Disulfide-linked dimers of human CD3~L appearunder non-reducing conditions, and even higher oligomers, apparently disulfide-linked, are seen with murine CD30-L. Most, but not all of these are converted to monomers upon reduction. The fart that not all oligomers were converted to monomers may S reflect either dif~erenial glycosylation and/or inefficient reduction.
~ '.
A soluble fusion protein comprising an antibody Fc region poly~ peptide joined through a peptide linker to the N-terminus ~f a fragment of the hu~nan CD3~L
ex~acellular domain was pr~duced and tested for biological activity as follows. ~NA
encoding a soluble human CD30-L pc~lypeptide compnsing amino acids 47 (Asp) ~o 215 (Asp) of Figure S was isolated and ampli~led by PCR. The PCR was conducted by conventional procedures, using as the 5' primer an oligonucleotide comprisingnucleotides 139-153 of Figure S and a sequence containing a recogninon site for ~15 BspE1. The 3' primer spanned the terminanon codon of CD3Q-L and contained the recognition sequence for Not I.
The PCR products were digested with Bsp E1 and Not I and the desired ~ragment was ligated into an expression vec~or designat~d pDC40~, which is a d~ivative of the pDC406 vector described above. pDC408 had been modified to contain (in order) 5'- murine IL-7 leader sequence - FLAG(~ - human lgG1 Fc domain - peptide linker.
The murine IL-7 leader sequence is described in U.S. Patent 4,~65,195 and the ~LAG(~ octapeptide is described above. The Fc polypeptide is descnbed in example 1.
A~peptide linker of the sequence Gly4SerGlysSer was employed, and the soluble ~5 CD30~L encoding DNA was inser~ed irnmediately downstream of the pep~ide linker, in ~he same reading ~ame. 293 cells (ATCC CRL 1573; a transformed primary human embryonal kidney cell line) were ~ansfected with the recombinant expr~ssion vector and cultured to pern~it expression and secre~on of the fusion protein. The expressed protein was pU~led on a pr~otein A column.
The activity of t}~e expressed pro~ein was measured using an inhibition assay inwhich the binding of 125I-la~elled CD30/Fc protein to CD30-L expressed on the sur~ace of trans~ormed CVl/E~NA cells was measu~ed. The soluble CD30-L-containing ~usion protem was shown to be capable of inhibiting dlis binding, thus indicating its abili~y ~o bind to ~30/Fc. The measured af~mi~y of the soluble ligand ~or CD30~Fc was roughly equivalent to that of GD30/Fc ~or the cell-bound ligand.

Claims (23)

What is claimed is:

1. An isolated DNA sequence encoding a biologically active CD30-L
polypeptide, wherein said CD30-L comprises an amino acid sequence selected from the group consisting of amino acids 1-220 of figure 3, amino acids 1-215 of figure 5, amino acids 1-239 of figure 6, and amino acids 1-234 of figure 7.

2. An isolated DNA sequence encoding a soluble CD30-L polypepnde, wherein said CD30-L comprises an amino acid sequence selected from the group consisting of amino acids 49-220 of figure 3 and amino acids z-215 of figure 5, wherein z is selected from the group consisung of 44, 45, 46, and 47.

3. An isolated DNA sequence according to claim 2, wherein said DNA
sequence additionally encodes an Fc polypeptide denved from an antibody fused, directly or through a peptide linker, to the N-terminus of the CD30-L polypeptide.

4. An isolated DNA capable of hybridizing to a DNA sequence of claim 1 under moderately stringent conditions, wherein said isolated DNA encodes a biologically active CD30-L.

5. An isolated DNA sequence according to claim 4, wherein said CD30-L
comprises an amino acid sequence selected from the group consisting of amino acids x to 239 of figure 6, wherein x is 1-19, and amino acids y to 234 of figure 7, wherein y is 1-19.

6. An expression vector comprising a DNA sequence according to claim 1.

7. An expression vector comprising a DNA sequence according to claim 2.

8. An expression vector comprising a DNA sequence according to claim 3.

9. An expression vector comprising a DNA sequence according to claim 4.

10. A process for preparing a CD30-L polypeptide, comprising culturing a host cell transformed with a vector according to claim 6 under conditions promoting expression of CD30-L, and recovering the CD30-L polypeptide.

11. A process for preparing a CD30-L polypeptide, comprising culturing a host cell transformed with a vector according to claim 7 under conditions promoting expression of CD30-L and recovering the CD30-L polypeptide.

12. A process for preparing a soluble CD30-L/Fc fusion protein, comprising culturing a host cell transformed with a vector according to claim 8 under conditions promoting expression of CD30-L/Fc, and recovering the CD30-L/Fc polypeptide.

13. A process for preparing a CD30-L polypeptide, comprising culturing a host cell transformed with a vector according to claim 9 under conditions promoting expression of CD30-L, and recovering the CD30-L polypeptide.

14. A substantially homogeneous purified biologically active CD30-L protein, wherein said CD30-L is selected from the group consisting of murine CD30-L
comprising the N-terminal amino acid sequence Met-Gln-Val-Gln-Pro-Gly-Ser-Val-Ala-Ser-Pro-Trp or Met-Glu-Pro-Gly-Leu-Gln-Gln-Ala-Gly-Ser-Cys-Gly, and human CD30-L comprising the N-terminal amino acid sequence Met-His-Val-Pro-Ala-Gly-Ser-Val-Ala-Ser-His-Leu or Met-Asp Pro-Gly-Leu-Gln-Gln-Ala-Leu-Asn-Gly-Met.

15. A punfied CD30-L according to claim 14, wherein said CD30-L comprises an amino acid sequence selected from the group consisting of amino acids 1-220 of figure 3, amino acids 1-215 of figure 5, amino acids 1-239 of figure 6, and amino acids 1-234 of figure 7.

16. A substantially homogeneous soluble CD30-1, polypeptide, wherein said soluble CD30-L comprises an amino acid sequence selected from the group consisting of amino acids 49-220 of figure 3 and amino acids z-215 of figure 5, wherein z is selected from the group consisting of 44, 45, 46, and 47.

17. Essentially homogeneous purified biologically active CD30-L protein, wherein said CD30-L is encoded by a DNA sequence that will hybridize to the nucleotide sequence presented in figure 3 or figure 5 under moderately stringentconditions.

18. Purified CD30-L according to claim 17, wherein said CD30-L comprises an amino acid sequence selected from the group consisting of amino acids x to 239 of figure 6, wherein x is 1-19, and amino acids y to 234 of figure 7, wherein y is 1-19.

19. A fusion protein comprising a CD30-L according to claim 17, wherein said CD30-L is a soluble CD30-L, and an Fc polypeptide derived from an antibody.

20. A dimeric protein comprising two fusion proteins according to claim 19, joined by disulfide bonds between the Fc polypeptides.

21. An antibody immunoreactive with CD30-L or an immunogenic fragment of CD30-L.

22. An antibody according to claim 21 wherein said antibody is a monoclonal antibody.

23. An antisense or sense oligonucleotide that can inhibit transcription or translation of CD30-L, comprising a sequence of at least about 14 nucleotides corresponding to a DNA sequence according to claim 1 or its DNA or RNA
complement.